IMAGE PROCESSING APPARATUS AND IMAGE PROCESSING METHOD
TECHl^ICAL FIELD [0001]
The present disclosure relates to an image processing apparatus and an image processingmethod and, more particularly, to an image processing apparatus and an image processingmethod capable of suppressing a deterioration in image guality.
BACKGROUND ART
[0002]
In recent years, apparatuses for compressing and encoding an image by employing an encoding scheme in v/hich image information is treated as digital data and, at this time, for the purpose of high-efficiency information transmission and accumulation, compression is performed through orthogonal transform such as discrete cosine transform and motion compensation by using redundancy unique to the image information have been v/idely used. The encoding scheme includes, for example, an MPEG (Moving Picture Experts Group) or the like.
[0003]
Particularly, the MPEG-2 (ISO/IEC 13818-2) scheme is defined as a general-purpose image encoding scheme and is a standard covering both of interlaced scanning images and sequential scanning images and covering standard resolution images and high-accuracy images. For example, the MPEG-2 scheme is v/idely used for a wide range of applications of professional uses and consumer uses. By using the MPEG2 compression scheme, for example, a code amount {bit rate) of

4 to 8 Mbps is allocated to an interlaced scanning image having a standard resolution of 720 x 480 pixels. In addition, by using the MPEG2 compression scheme, for example, a code amount
{bit rate} of 18 to 22 Mbps is allocated to an interlaced scanning image having a high resolution of 1920 x 1088 pixels. Therefore, a high compression rate and a good image quality may be implemented.
[0004]
The MPEG-2 is mainly applied to high image quality encoding v/hich is suitable for broadcasting, but it does not correspond to an encoding scheme having a code amount (bit rate) lov/er than that of the MPEGl, that is, an encoding scheme having a higher compression rate. With the spread of mobile phones, needs for the encoding scheme are expected to be increased, and accordingly, the MPEG-4 encoding scheme is standardized. With respect to the image encoding scheme, the ISO/IEC 14496-2 standard was approved as an international standard in December, 1998.
[0005]
In addition, in recent years, for the purpose of image encoding for TV conference, standardization called H.26L
(ITU-T (International Telecommunication Union Telecommunication Standardization Sector) Q6/16 VCEG {Video Coding Expert Group)) has been promoted. It is knov;n that, in comparison v/ith the encoding schemes such as the MPEG~2 or the MPEG-4 in the related art, in the H.26L, although a large calculation amount is needed for encoding and decoding, a higher encoding efficiency is implemented. In addition, at present, as a part of activities of the MPEG-4, standardization v/hich is based on the H.26L and incorporates functions v/hich are not supported in the H.26L to implement

a higher encoding efficiency is performed as Joint Model of
Enhanced-Compression Video Coding.
[0006]
As the schedule of the standardization, the standard was approved as an international standard on the basis named H.264 and MPEG-4 Part 10 (Advanced Video Coding, hereinafter referred to as AVC)) in March, 2003. [0007]
In addition, as extension of the H.264/AVC, standardization of Fidelity Range Extension (FRExt) including RGB, encoding tools necessary for business such as 4:2:2 or 4:4:4, 8x8 DCT defined by the MPEG-2, and quantization matrices v/as completed in February, 2005. Accordingly, the H.264/AVC became an encoding scheme capable of representing film noise included in a movie with a good quality. Therefore, the H.264/AVC has been used for a v?ide range of applications such as Blu-Ray disc (trade marJc) . [0008]
Hov?ever, recently, needs for high compression rate encoding, for example, a need to compress images of about 4000 X 2000 pixels v/hich is four times of a high-vision image or a need to distribute a high-vision image in a limited-transmission-rate environment such as the Internet have been further increased. Therefore, in the VCEG under the ITU-T, improvement of an encoding efficiency continues to be studied. [0009]
Therefore, at present, for the purpose of further improvement of the encoding efficiency in comparison v/ith the AVC, standardization of an encoding scheme called high efficiency video coding (HEVC) has been promoted by the joint

collaboration team-video coding {JCTVC) as a joint standardization body of the ITU-T and the ISO/IEC. With respect to the HEVC standard, Committea Draft as a first draft specification v/as issued in February, 2012 (for example, refer to Non-Patent Document 1).
[0010]
Hov/ever, in an intra 8x8 prediction mode of the AVC encoding scheme, a [121]/4 filtering process is performed on neighboring pixels of a current blocJ^ v;hich is a processing target. In the HEVC, on/off of the filtering process is determined according to a block size and a prediction mode.
[0011]
In addition, in the HEVC, for the purpose of reducing blocJ^ distortion of the case v/here the prediction mode is a DC mode, a horizontal mode, or a vertical mode, a boundary value smoothing process is defined.
[0012]
In addition, in order to reduce a phenomenon that a contour is seen at a flat portion of an image, a contour noise countermeasure process was proposed (for example, refer to Non-Patent Document 2).
[0013]
In the method disclosed in Non-Patent Document 2, features of the neighboring pixels of the current block v/hich is a processing target are identified by performing a threshold value determining process. In the case where a result of the threshold value determining process is true, namely, in the case where theneighboringpixels have predetermined features, instead of the above-described [121]/4 filtering process, a bi-linear interpolation process (referred to as a bi-linear filtering process) is performed.

CITATION LIST NON-PATENT DOCDMENT [00141
Non-Patent Document 1: Benjamin Bross, Woo-Jin Han^ Jens-Rainer Ohm, Gary J. Sullivan, Thomas Wiegand, "High efficiency video coding (HEVC) text specification draft 8", JCTVC-Hl003_d7, Joint Collaborative Team on Video Coding (JCT-VC) of ITQ-T SG16WP3 arid ISO/IEC JTC1/SC29/WG11, 10th Meeting; Stockholm, SE, 11-20 July 2012
Non-Patent Document 2 : TKTan, Y. Suzuki, " Contouring artefact and solution", JCTVC-K0139, Joint Collaborative Team on Video Coding (JCT-VC) of ITU-T SG 16WP3 and ISO/IEC JTC 1/SC 29/WG 1111th Meeting; Shanghai,

CN, 10-19 Oct. 2012 SUMMARY OF THE INVENTION PROBLEMS TO BE SOLVED BY THE INVENTION
[0015]
Hov/ever, in the method of disclosed in Non-Patent Document 2, the threshold value is fixed. Namely, irrespective of the feature, for example, the bit depth or the like of the image vjhich is a target of the encoding/decoding process, the same value is set as the threshold value. Therefore, there is a problem in that, in the selection of the filtering process on the neighboring pixels in the intra prediction, an appropriate filter is not selected, so that image quality is unnecessarily deteriorated.
[0016]
The present technigue is to suppress a deterioration in image guality.
SOLUTIONS TO PROBLEMS [0017]
According to an aspect of the present technigue, there is provided an image processing apparatus including a threshold value setting unit which sets a threshold value which is compared viith a value calculated by using a sample of neighboring pixels in order to identify a feature of the neighboring pixels of a current block in an intra prediction process in encoding of image data according to a bit depth of the image data and a filtering process unit which performs a filtering process on the neighboring pixels by using a bi-linear interpolation filter in the case where the value calculated by using the sample of the neighboring pixels is lov/er than the threshold value set by the threshold value

setting unit. [0018]
The threshold value setting unit may bit-shift the threshold value which is determined as an initial value in advance according to the bit depth. [0019]
The threshold value setting unit may set the threshold value to 8 in the case where the bit depth of the image data is 8 bits. [0020]
The image processing apparatus may further include a transmitting unit vrhich transmits the threshold value set by the threshold value setting unit. [0021]
The image processing apparatus may further include a determining unit v/hich determines the bit depth of the image data/ and the threshold value setting unit may set the threshold value according to the bit depth determined by the determining unit. [0022]
The image processing apparatus may further include a transmitting unit v/hich transmits the bit depth determined by the determining unit. [0023]
The filtering process unit may perform a low pass filtering process on the neighboring pixels in the case vzhere the value calculated by using the sample of the neighboring pixels exceeds the threshold value set by the threshold value setting unit.

According to another aspect of the present technigue, there is provided an image processing method including setting a threshold value v/hich is compared with a value calculated by using a sample of neighboring pixels in order to identify a feature of the neighboring pixels of a current block in an intra prediction process in encoding of image data according to a bit depth of the image data and performing a filtering process on the neighboring pixels by using a bi-linear interpolation filter in the case v/here the value calculated by using the sample of the neighboring pixels is lower than the set threshold.
[00251
In the aspect of the present technique, a threshold value v/hich is compared v/ith a value calculated by using a sample of neighboring pixels in order to identify a feature of the neighboring pixels of a current block in an intra prediction process in encoding of image data is set according to a bit depth of the image data, and a filtering process is performed on the neighboring pixels by using a bi-linear interpolation filter in the case where the value calculated by using the sample of the neighboring pixels is lov/er than the set threshold.
EFFECTS OF THE INVENTION [002 Q
According to the present technique, it is possible to encode and decode an image. Particularly, it is possible to suppress a deterioration in image quality.
BRIEF DESCRIPTION OF DRAWINGS [0028]

Fig. 1 is a diagram illustrating a configuration example of a coding unit.
Fig. 2 is a diagram illustrating an example of angular prediction.
Fig. 3 is a diagram illustrating an example of planar prediction.
Fig. 4 is a diagram illustrating an example of a most probable mode.
Fig. 5 is a diagram illustrating an example of MDIS (Mode Dependent Intra Smoothing).
Fig. 6 is a diagram illustrating an example of a boundary value smoothing process.
Fig. 7 is a diagram illustrating an example of a decoded image.
Fig. 8 is a diagram illustrating another example of a decoded image.
Fig . 9 is a diagram illustrating an example of a behavior of a threshold value determining process.
Fig. 10 is a block diagram illustrating a main configuration example of an image encoding device.

Fig. 11 is a block diagram illustrating a main configuration example of a threshold value setting unit and filtering processing unit.
Fig. 12 is a flowchart illustrating an example of a flow of an encoding process.
Fig. 13 is a flowchart illustrating an example of a flow of a threshold value setting process.
Fig. 14 is a flowchart illustrating an example of a flow of an intra prediction process.
Fig. 15 is a flov;chart illustrating another example of a flow of a threshold value setting process.
Fig. 16 is a f lov/chart illustrating still another example of a f lov^ of a threshold value setting process.
Fig. 17 is a block diagram illustrating another configuration example of a threshold value setting unit and a filtering processing unit.
Fig. 18 is a f lovjchart illustrating still another example of a flovj of a threshold value setting process.
Fig. 19 is a block diagram illustrating still another configuration example of a threshold value setting unit and a filtering processing unit.
Fig . 20 is a f lov;chart illustrating still another example of a flow of a threshold value setting process.
Fig . 21 is a flowchart illustrating still another example of a flow of a threshold value setting process.
Fig. 22 is a flovjchart illustrating still another example of a flow of a threshold value setting process.
Fig. 23 is a block diagram illustrating a main configuration example of an image decoding device.
Fig. 24 is a block diagram illustrating a main configuration example of a threshold value setting unit and

a filtering processing unit.
Fig. 25 is a flov/chart illustrating an example of a flovj of a decoding process.
Fig. 26 is a flov/chart illustrating an example of a flow of a threshold value setting procesa.
Fig. 27 is a flowchart illustrating an example of a flow of a prediction process.
Fig. 28 is a flowchart illustrating an example of a flow of an intra prediction process.
Fig. 29 is a flovjchart illustrating another example of a flov? of a threshold value setting process.
Fig, 30 is a block diagram illustrating another configuration example of a threshold value setting unit and a filtering processing unit.
Fig. 31 is aflowchart illustrating still another example of a flov? of a threshold value setting process.
Fig. 32 is a flov/chart illustrating still another example of a flow of a threshold value setting process.
Fig. 33 is a block diagram illustrating still another configuration example of a threshold value setting unit and a filtering processing unit.
Fig . 34 is a flowchart illustrating still another example of a flov/ of a threshold value setting process.
Fig. 35 is a block diagram illustrating still another configuration example of a threshold value setting unit and a filtering processing unit.
Fig. 3 6 is a flowchart illustrating still another example of a flow of a threshold value setting process.
Fig. 37 is a diagram illustrating an example of a multi-viev/point image encoding scheme.
Fig. 38 is a diagram illustrating amain configuration

example of a multi-viewpoint image encoding device to v/hich the present technique is applied.
Fig. 39 is a diagram illustrating a main configuration example of a multi-viewpoint image decoding device to v/hich the present technique is applied.
Fig. 40 is a diagram illustrating an example of a hierarchical image encoding scheme.
Fig. 41 is a diagram illustrating an example of spatial scalable encoding.
Fig. 42 is a diagram illustrating an example of temporal scalable encoding.
Fig. 43 is a diagram illustrating an example of signal-to-noise ratio scalable encoding.
Fig. 44 is a diagram illustrating a main configuration example of a hierarchical image encoding device to which the present technique is applied.
Fig. 45 is a diagram illustrating a main configuration example of a hierarchical image decoding device to which the present technique is applied.
Fig. 46 is a block diagram illustrating a main configuration example of a computer.
Fig. 47 is a block diagram illustrating an example of a schematic configuration exarrple of a television apparatus.
Fig. 48 is a block diagram illustrating a schematic configuration example of a mobile phone.
Fig. 49 is a block diagram illustrating a schematic configuration exarrple of a recording/reproducing apparatus.
Fig. 50 is a block diagram illustrating a schematic configuration example of an imaging apparatus.
Fig. 51 is a block diagram illustrating an example of use of scalable encoding.

Fig. 52 is a block diagram illustrating another example of use of scalable encoding.
Fig. 53 is a block diagram illustrating still another example of use of scalable encoding.
Fig. 54 is a block diagram illustrating a schematic configuration example of a video set.
Fig. 55 is a block diagram illustrating a schematic configuration example of a video processor.
Fig. 56 is a block diagram illustrating another schematic configuration example of a video processor.
Fig. 57 is a diagram illustrating a configuration of a content reproducing system.
Fig. 58 is a diagram illustrating a flov; of data in a content reproducing system.
Fig . 59 is a description diagram illustrating a specific example of an MPD.
Fig. 60 is a functional block diagram illustrating a configuration of a content server of a content reproducing system.
Fig. 61 is a functional block diagram illustrating a configuration of a content reproducing apparatus of a content reproducing system.
Fig. 62 is a functional block diagram illustrating a configuration of a content server of a content reproducing system.
Fig. 63 is a sequence chart illustrating an example of a communication process of each apparatus in a v/ireless communication system.
Fig. 64 is a sequence chart illustrating an example of a communication process of each apparatus in a v/ireless communication system.

Fig. 65 is a schematic diagram illustrating a configuration example of a frame format transmitted/received in a communication process by each apparatus in a v/ireless communication system.
Fig. 66 is a sequence chart illustrating an example of a communication process of each apparatus in a v/ireless communication system.
MODE FOR CARRYING OUT THE INVENTION [0029]
Hereinafter, embodiments for implementing the present disclosure {hereinafter, referred to as embodiments) will be described. In addition, the description is performed in the following order.
0. Overviev/
1. First Embodiment {Image Encoding Device)
2. Second Embodiment {Image Decoding Device)
3. Third Embodiment {Multi-Viev/point Image Encoding/Multi-Viev/point Image Decoding Device)
4. Fourth Embodiment {Hierarchical Image Encoding ■ Hierarchical Image Decoding Device)
5. Fifth Embodiment (Computer)
6= Example of Application
7. Example of Application of Scalable Encoding.
8. Sixth Embodiment (Set/Unit/Module/Processor)
9. Seventh Embodiment (Example of Application of Content
Reproducing System of MPEG-DASH)
10. Eighth Embodiment (Example of Application of
Wireless Communication System of Wi-Fi Standard)
[0030]
<0. Overview>

w^^

SP351032WO001

Hereinafter, the present technique v?ill be described in an example where the present technique is applied to imaqe encodinq/decodinq in an HEVC (High Efficiency Video Codinq) 5 scheme. [0031]

IntheAVC (AdvancedVideoCodinq) scheme, ahierarchical structure having macroblocks and submacroblocks are defined. 10 Hov^ever, the macroblock of 16 x 16 pixels is not optimal to a larqe imaqe frame called UHD (Ultra Hiqh Definition, 4000 X 2000 pixels) , which is to be an object of a next-qeneration encodinq scheme. [0032] 15 On the other hand, in the HEVC scheme, as illustrated in Fiq. 1, a codinq unit (CU) is defined. [0033]
The CU is also called a codinq tree block (CTB) and is a partial reqion of an imaqe in units of a picture v;hich has 20 the same functions as that of the macroblock in the AVC scheme . The latter is fixed in size of 16 x 16 pixels, but the former is not fixed in size but it is desiqnated in imaqe compression information in each sequence. [0034] 25 For example, in a sequence parameter set (SPS) included in output encoded data, a maximum size (LCU (Larqest Codinq unit)) of the CU and a minimum size (SCU (Smallest Codinq Unit) ) of the CU are defined. [0035] 30 In each LCU, the CU can be split into CUs havinq a smaller size by settinq split-flaq = 1 v/ithin a ranqe where the size

SP351032WO001
of CU is not smaller than the size of SCU. In the example of Fig. 1, the size of LCU is 128, and the maximum hierarchical depth is 5. When the value of split__flag is "1", the CU having a size of 2N x 2N is split into the CUs having a size of N 5 X N v/hich are disposed in the one lower layer. [0036]
In addition, the CU is split into prediction units (PUs) as the regions (partial regions of an image in units of a picture) v/hich become a processing unit of intra or inter prediction,
10 or the CU is split into transform units (TUs) as the regions (partial regions of an image in units of a picture) v/hich become a processing unit of orthogonal transform. At present, in the HEVC scheme, in addition to 4 x 4 orthogonal transform and 8x8 orthogonal transform, 16 x 16 orthogonal transform
15 and 32 x 32 orthogonal transform can be used. [0037]
Like the above-described HEVC scheme, it is considered that, in the encoding scheme where the CU is defined and various processes are performed in units of the CU, a macroblock in
20 the AVC scheme corresponds to the LCU and a block (subblock) corresponds to the CU. In addition, it is considered that a motion compensation block in the AVC scheme corresponds to the PU. Hov/ever, since the CU has a hierarchical structure, asize (forexample, 128 x 128 pixels) of the LCU of the uppermost
25 layer is generally set to be larger than the size of the macroblock of the AVC scheme. [0038]
Therefore, hereinafter, the LCU is assumed to include the macroblock in the AVC scheme, and the CU is assumed to
30 include the block (subblock) of the AVC scheme. Namely, the "block" used for the f ollov^ing description denotes anarbitrary

SP351032WO001
partial area within a picture, and a size, a shape, a feature, and the like thereof are not limited. Namely, the "block" includes, for example, the TU, the PU, the SCU, the CU, the LCU, the subblock, the macroblock, or an arbitrary area 5 (process unit) such as a slice. In addition, other partial areas (process unit) are also included. In the case v/here the size, the process unit, or the like needs to be limited, appropriate description thereof v/ill be made. [0039]
10
However, in order to achieve a higher encodingef f iciency in the AVC or HEVC encoding scheme, it is important to select an appropriate prediction mode. [0040]
15 As an example of the associated selection scheme, there is a method incorporated in reference software (disclosed in http://iphome.hhi.de/suehring/tml/index.htm) of H.264/MPEG~4AVC, v/hich is called JM (Joint Model). [0041]
20 In the JM, tvfo mode determining methods of high
complexity mode and lov; complexity mode described belovj can be selected. In both of the methods, cost function values with respect to the respective prediction mode modes are calculated, and the prediction mode v/here the cost function
25 value is minimized is selected as an optimal mode for an associated block or macroblock. [0042]
The cost function in the high complexity mode is expressed by the follovjing Formula (1) .
30 [0043]
[Mathematical Formula 1]

=1

SP351032WO001
Cost(Mode G H) - D + X*R ... (1)
[0044]
Here, VL denotes a total set of candidate modes for
encoding the associated block or macroblock, and D denotes 5 difference energy betv^een a decoded image and an input image
in the case of encoding the associated prediction mode. %
denotes a Lagrange multiplier given as a function of a
quantization parameter. R denotes a total code amount
including an orthogonal transform coefficient in the case of 10 encoding the associated mode.
[0045]
Namely, v/hen encoding in a High complexity mode is
intended to be performed, in order to calculate the
above-described parameters D and R, a preliminary encoding 15 process needs to be performed once in every candidate mode,
so that a higher calculation amount is reguired.
[0046]
The cost function in the lov/ complexity mode is expressed
by the following Formula (2). 20 [0047]
[Mathematical Formula 2]
Cost(Mode 6 H) = D + QP2Quant(QP)*HeaderBit ... (2) [0048]
Here, unlike the case of the high complexity mode, D
25 becomes difference energy betv/een a predicted image and an input image. QP2Quant (QP) is given as a function of a quantization parameter QP, HeaderBit is a code amount vzith respect to information included in Header, vzhich does not include the orthogonal transform coefficient and is called
30 a motion vector or a mode. [0049]

SP351032WO001
Namely, in the lov/ complexity mode, although the prediction process in each candidate mode needs to be performed, since a decoded image is not needed, the encoding process needs not be performed. Therefore, it is possible to implement v/ith 5 a lov/er calculation amount than that in the high complexity mode. [0050]

In the AVC, intra 4x4 prediction, intra 8x8 prediction, 10 and intra 16 x 16 prediction exist. In the HEVC, as illustrated in Fig. 2, angular prediction is applied to 4 x 4 to 64 x 64 pixel blocks. [0051]
Namely, in the AVC, as illustrated in A of Fig. 2, the 15 intra prediction process is performed by 8-direction -i- DC prediction. In contrast, in the HEVC, as illustrated in B of Fig. 2, the intra prediction process is performed by 32-direction + DC prediction. Accordingly, the prediction accuracy is improved. 20 [0052]
In addition, in the HEVC, as illustrated in Fig . 3, planar prediction is defined. [0053]
In the planar prediction process, prediction pixels 25 included in the current block are generated from neighboring pixels (pixels that are already encoded) of a current block v/hich is a target of the process by bi-linear interpolation. The planar prediction process can improve encoding efficiency for an area v/here there is gradation. 30 [0054]
In the HEVC, as illustrated in Fig. 4, the encoding

-i^To

SP351032WO001

process is performed in the intra prediction mode using three most probable modes . Namely^ an intra prediction mode (Above) of a neighboring block v;hich is above the current block to be adjacent to the current block, an intra prediction mode 5 (Left) of a neighboring block which is in the left of the current block to be adjacent to the current blocks and a combination of the intra prediction modes of the neighboring blocks (Above and Left) are used as candidates (sometimes, referred to as candidate modes) of the intra prediction mode, and among the 10 three candidates modes, the optimal mode is employed as the intra prediction mode for the current block. [0055]
If the prediction mode for the current block is equal to any of the most probable modes, the index number is 15 transmitted. If not, the mode information of the prediction block is transmitted by a fixed length of 5 bits. [0056]

Fig. 5 is a diagram describing MDIS (Mode Dependent Intra 20 Smoothing) regulated in the HEVC. [0057]
In the case of the AVC, as illustrated in Fig . 5, a [121]/4 filtering process is performed on neighboring pixels of a current block in an intra 8x8 prediction mode. On the other 25 hand, in the HEVC, on/off of the filtering process (namely, v/hether or not to apply the filtering process) is decided according to the block size and the prediction mode. [0058]
More specifically, in the case where the block size of
30 the current block is 4 x 4, the filtering process is not applied.
In the case vj^here the block size of the current block is 8

SP351032WO001
X 8, in a prediction mode of the 45-degree direction, the filtering process is applied. In the case v/here the block size of the current block is 16 x 16, in a prediction mode of a direction other than 3 directions close to the horizontal 5 direction and 3 directions close to the vertical direction, the filtering process is applied. In the case where the block
size of the current block is 32 x 32, in a prediction mode of a direction other than the horizontal directions and the vertical direction, the filtering process is applied.
10 [0059]
Furthermore, in the HEVC, for the purpose of reducing block distortion in the case v/here the prediction mode is the DC mode, the horizontal mode, or the vertical mode, a boundary-value smoothing process illustrated in Fig. 6 is regulated.
15 [0060]
For example, in the case of the prediction (DC prediction) v/here the prediction mode is the DC mode, the filtering process (smoothing process) illustrated in Fig. 6 is performed on both of the neighboring pixels adjacent to
20 the upper side (Top) of the current block v/hich is a target of the process and the neighboring pixels adjacent to the left side (Left) of the current block. In addition, in the case of the prediction (hori zontal prediction) v/here the prediction mode is the horizontal mode, the filtering process (smoothing
25 process) illustrated in Fig . 6 is performed on the neighboring pixels adjacent to the upper side (Top) of the current block. In the case of the prediction (vertical prediction) v/here the prediction mode is the vertical mode, the filtering process (smoothing process) illustrated in Fig. 6 is performed on the
30 neighboring pixels adjacent to the left side (Left) of the current block.

-1 --

SP351032WO001
[0061]
In the case v/here the image is encoded/decoded by the above-described inter prediction, in the obtained decoded image, there is a portion vjhere the change of luminosity, color, 5 density or the like is planar as illustrated in the area 11 of Fig. 7, and a strip shape of density irregularity (so-called bending) occurs, so that the phenomenon v^here contour is seen may occur. [0062]
10 Therefore, a contour noise countermeasure process disclosed in Non-Patent Document 2 v^as proposed. By performing the contour noise countermeasure process disclosed in Non-Patent Document 2, the bending occurring in the area 11 of the decoded image is suppressed as illustrated in Fig.
15 8, so that smooth gradation can be obtained. [0063]
The contour noise countermeasure process disclosed in Non-Patent Document 2 is described more in detail. Fig. 9 is adiagramillustratinganexampleof a behavior of the contour
20 noise countermeasure process. In the contour noise countermeasure process, firstly, a threshold value determining process expressed by the following Formulas (3) and (4) is performed by using neighboring pixels of a current block illustrated in Fig. 9.
25 [0064]
[Mathematical Formula 3]
Interpolate_Above = abs(AL + AR - 2*ref[3N]} <
THRESHOLD_AB0VE ... (3)
Interpolate_Left = abs{BL + AL - 2*ref[N]) <
30 THRESHOLD__LEFT ... (4) [0065]

SP351032WO001
In Formula (3), the value of the threshold value THRESHOLD is set to be fixed to 8. By the threshold value determining process, features of the neighboring pixels of the current block is determined. Namely, it is determined 5 v/hether or not the periphery of the current block is a portion where the change of luminosity, color, density, or the like is planar so that the bending can occur. In the case where the result of the threshold value determining process is true, that is, in the case v/here it is determined that the periphery 10 of the current block is a portion v/here the change of luminosity, color, density, or the like is planar so that the bending can occur, instead of the filtering process described viith reference to Fig. 5, a bi-linear interpolation process expressed by the follov/ing Formulas (5) to (9) is performed 15 on the neighboring pixels of the current block illustrated in Fig. 9. [0066]
[Mathematical Formula 4] ref'[0] = ref[0] ... (5) 20 ref [i] = BL + i* {AL - BL + N)/2N (i = 1 to 2N - 1) (6)
ref [2N] = ref [2N] ... (7)
ref [2N -f i] =AL + i* (AR-AL + N)/2N (i = 1 to 2N - 1) (8) 25 ref [4N] = ref [4N] ... (9) [0067]
The process is applied to only the 32 x 32 block, and a flag representing whether or not to apply the process (on/off) is regulated in a sequence parameter set (SPS). 30 [0068]
However, in the method disclosed in Non-Patent Document

t

3^'>H

SP351032WO001

2, the threshold value v/as set to be fixed. Therefore, there v/as a problem in that, in the selection of the filtering process on the neighboringpixels in the intra prediction, appropriate selection is not performed, so that image quality is 5 unnecessarily deteriorated. [0069]
In the threshold value determining process, as expressed in Formulas (3) and (4), a values calculated from pixels value of the neighboringpixels of the current block and the threshold
10 value are compared with each other . Hov/ever, if the threshold value is set to be fixed, in some cases, it is considered that, the value is not appropriate to, for example, the bit depth of the image data is considered. In this case, there is a problem in that, as a result of the above-described threshold
15 value determining process, a proper determination result is not obtained, and an appropriate filter is not selected, so image quality of the decoded image is unnecessarily deteriorated. [0070]
20 In addition, it is considered that, for example, the bit depth of the image data is changed in internal calculation at the time of encoding/decoding. In this case, similarly, there is a problem in that a fixed threshold value is not appropriate to the bit depth in the internal calculation, and
25 an improper result of the threshold value determining process is obtained, so that image quality of the decoded image is unnecessarily deteriorated. [0071]
In addition, the image quality of the decoded image
30 cannot be adjusted, for example, by the user or the like adjusting the threshold value.


SP351032WO001
[0072]
Therefore, in the present technigue, the threshold value
is set to be variable. For example, the threshold value is
allowed to be set according to the bit depth of the image data. 5 By doing so, it is possible to suppress a deterioration in
image guality. In addition, by setting the threshold value
to be variable, it is possible to adjust the image guality
of the decoded image.
[0073] 10 Next, with respect to the above-described present
technigue, examples of application to specific devices v;ill
be described.
[0074]
<1. First Embodiment> 15
Fig. 10 is a block diagram illustrating a configuration
example of an image encoding device as a kind of an image
processing apparatus to v;hich the present technigue is applied.
The image encoding device 100 illustrated in Fig. 10 encodes 20 image data of a moving image by using, for example, a prediction
process of the HEVC or a prediction process in a scheme
eguivalent to the HEVC.
[0075]
As illustrated in Fig. 10, the image encoding device 25 100 is configured to include an A/D converter 101, a screen
rearrangement buffer 102, an arithmetic unit 103, an orthogonal
transform unit 104, a guantization unit 105, a lossless
encoding unit 106, an accumulation buffer 107, an inverse
guantization unit 108, and an inverse orthogonal transform 30 unit 109. In addition, the image encoding device 100 is
configured to include an arithmetic unit 110, a loop filter

SP351032WO001
111, a frame memory 112, an intra prediction unit 113, an inter prediction unit 114, a predicted image selection unit 115, and a rate control unit 116. [0076] 5 The A/D converter 101 A/D-converts input image data and supplies the converted image data (digital data) to the screen rearrangement buffer 102 to store the converted image data. The screenrearrangementbuf fer 102 rearranges the images v;hich are in the stored frame order for display by using the frame
10 order for encoding according to a GOP (Group of Picture) and supplies the images of v/hich frame order is rearranged to the arithmetic unit 103. In addition, the screen rearrangement buffer 102 also supplies the images of v/hich frame order is rearranged to the intra prediction unit 113 and the inter
15 prediction unit 114. [0077]
The arithmetic unit 103 subtracts the predicted image supplied from the intra prediction unit 113 or the inter prediction unit 114 through the predicted image selection unit
20 115 from the image read from the screen rearrangement buffer 102 and outputs difference information thereof to the orthogonal transform unit 104. For example, in the case of an image on which intra encoding is performed, the arithmetic unit 103 subtracts the predicted image supplied from the intra
25 prediction unit 113 from the image read from the screen
rearrangement buffer 102. In addition, for example, in the case of an image on which inter encoding is performed, the arithmetic unit 103 subtracts the predicted image supplied from the inter prediction unit 114 from the image read from
30 the screen rearrangement buffer 102. [0078]

SP351032WO001
The orthogonal transform unit 104 performs orthogonal transform such as discrete cosine transform or Karhunen-Loeve on the difference information supplied from the arithmetic unit 103. The orthogonal transform unit 104 supplies a 5 transform coefficient thereof to the quantization unit 105. [0079]
The quantization unit 105 performs quantization on the transform coefficient supplied from the orthogonal transform unit 104. The quantization unit 105 sets a guantization
10 parameter based on information on a target value of the code amount supplied from the rate control unit 116 and performs guantization thereof. The guantization unit 105 supplied the quantized transform coefficient to the lossless encoding unit 106.
15 [0080]
The lossless encoding unit 106 encodes the transform coefficient guantized in the guantization unit 105 in an arbitrary encoding scheme. Since the coefficient data are quantized under the control of the rate control unit 116/ the
20 code amount becomes the target value set by the rate control unitll6 (or the code amount isapproximate to the target value} . [0081]
In addition, the lossless encoding unit 106 acquires the information representing the intra prediction mode or the
25 like from the intra prediction unit 113 and acquires
information representing an inter prediction mode, the difference motion vector information, or the like from the inter prediction unit 114. [0082]
30 The lossless encoding unit 106 encodes the various kinds of information in an arbitrary encoding scheme to be used as

SP351032WO001
a portion of header information of the encoded data (sometimes, referred to as an encoded stream) . The lossless encoding unit 106 supplies the encoded data obtained through the encoding to the accumulation buffer 107 to accumulate the encoded data . 5 [0083]
The encoding scheme of the lossless encoding unit 106 includes, for example, variable length encoding, arithmetic encoding, and the like. The variable length encode includes, for example, CAVLC {Context-Adaptive Variable Length Coding)
10 defined in the H.2 64/AVC scheme and the like. The arithmetic encode includes, for example, CABAC (Context-Adaptive Binary Arithmetic Coding) and the like. [0084]
The accumulation buffer 107 temporarily stores the
15 encoded data supplied from the lossless encoding unit 106. The accumulation buffer 107 outputs the stored encoded data to an outside of the image encoding device 100 at a predetermined timing. Namely, the accumulation buffer 107 is also a transmitting unit which transmits the encoded data.
20 [0085]
In addition, the transform coefficient quantized in the quantization unit 105 is also supplied to the inverse quantization unit 108. The inverse quantization unit 108 performs inverse quantization on the quantized transform
25 coefficient in a method corresponding to the quantization of the quantization unit 105 . The inverse quantization unit 108 supplies the obtained transform coefficient to the inverse orthogonal transform unit 109. [0086]
30 The inverse orthogonal transform unit 10 9 performs inverse orthogonal transform on the transform coefficient

SP351032WO001
supplied from the inverse quantization unit 108 in a method corresponding to the orthogonal transform process of the orthogonal transform unit 104. The
inverse-orthogonal-transformed output (restored difference 5 information) is supplied to the arithmetic unit 110. [0087]
The arithmetic unit 110 adds the predicted image supplied from the intra prediction unit 113 or the inter prediction unit 114 through the predicted image selection unit 115 to
10 the recovered difference information v/hich is the inverse orthogonal transform result supplied from the inverse orthogonal transform unit 109 to obtain a locally recovered image {hereinafter, referred to as a reconstructed image). The reconstructed image is supplied to the loop filter 111
15 or the intra prediction unit 113. [0088]
The loop filter 111 appropriately performs a filtering process including a deblocking filter, an adaptive loop filter, or the like on the reconstructed image supplied from the
20 arithmetic unit 110. For example, the loop filter 111 removes block distortion of the reconstructed image by performing the deblocking filtering process on the reconstructed image- In addition, for example, the loop filter 111 improves the image quality by performing the loop filtering process on the
25 deblocking filtering process result {the reconstructed image from which the block distortion is removed) by using a Wiener Filter. [0089]
In addition, the loop filter 111 may further perform
30 any other arbitrary filtering process on the reconstructed image. In addition, if necessary, the loop filter 111 may

/

/

SP351032WO001
supply information such as a filter coefficient v/hich is used for the filtering process to the lossless encoding unit 106, so that the information may be encoded. [0090] 5 The loop filter 111 supplies the f ilteringprocess result (hereinafter, referred to as a decoded image) to the frame memory 112. [0091]
The frame memory 112 stores the supplied decoded image, 10 and at a predetermined timing, the frame memory supplies the stored decoded image as a reference image to the inter prediction unit 114. [0092]
The intra prediction unit 113 performs intra prediction 15 (prediction within a screen) of generating the predicted image by using pixel values in the process target picture v/hich is the reconstructed image supplied as the reference image from the arithmetic unit 110. The intra prediction unit 113 performs the intra prediction in a plurality of predetermined .20 intra prediction modes. [0093]
The intra prediction unit 113 generates the predicted images in all the intra prediction modes v/hich are candidates and evaluates the cost function values of the respective 25 predicted images by using the input image supplied from the screen rearrangement buffer 102 to select the optimal mode. Once the intra prediction unit 113 selects the optimal intra prediction mode, the intra prediction unit supplies the predicted image generated in the optimal mode to the predicted 30 image selection unit 115. [0094]

1

SP351032WO001
In addition, as described above, the intra prediction unit 113 appropriately supplies intra prediction mode information representing the selected intra prediction mode or the like to the lossless encoding unit 106, so that encoding 5 is performed. [0095]
The inter prediction unit 114 performs an inter prediction process (motion prediction process and motion compensation process) by using the input image supplied from
10 the screen rearrangement buffer 102 and the reference image supplied from the frame memory 112 , More specifically, the inter prediction unit 114 performs the motion compensation process according to the motion vector detected by performing the motion prediction as the inter prediction process to
15 generate the predicted image (Inter predicted image
information). The inter prediction unit 114 performs the inter prediction in a plurality of predetermined inter prediction modes. [0096]
20 The inter prediction unit 114 generates the predicted images in all the inter prediction modes which are candidates . The interprediction unit 114 evaluates the cost function value of each predicted image by using the input image supplied from the screen rearrangement buffer 102, the information of the
25 generated difference motion vector, and the like to select the optimal mode. Once the inter prediction unit 114 selects the optimal inter prediction mode, the inter prediction unit supplies the predicted image generated in the optimal mode to the predicted image selection unit 115.
30 [0097]
The inter prediction unit 114 supplies the information

SP351032WO001
representing the selected inter prediction mode or the information necessary for performing the process in the inter prediction mode at the time of decoding the encoded data to the lossless encoding unit 106, so that the encoding is 5 performed. The necessary information includes, for example, the information of the generated difference motion vector, a flag representing an index of as the prediction motion vector as the prediction motion vector information, and the like. [0098]
10 The predicted image selection unit 115 selects the supply source of the predicted image v/hich is to be supplied to the arithmetic unit 103 or the arithmetic unit 110. For example, in the case of the intra encoding, the predicted image selection unit 115 selects the intra prediction unit 113 as the supplying
15 source of the predicted image and supplies the predicted image supplied from the intra prediction unit 113 to the arithmetic unit 103 or the arithmetic unit 110 . In addition, for example, in the case of the inter encoding, the predicted image selection unit 115 selects the inter prediction unit 114 as the supplying
20 source of the predicted image and supplies the predicted image supplied from the inter prediction unit 114 to the arithmetic unit 103 or the arithmetic unit 110. [0099]
The rate control unit 116 controls a rate of guantization
25 operation of the guantization unit 105 based on the code amount of the encoded data accumulated in the accumulation buffer 107 so that overflov/ or underflow does not occur. [0100]
The image encoding device 100 is configured to further
30 include a threshold value setting unit 121 and a filtering processing unit 122.

"/

SP351032WO001
[0101]
The threshold value setting unit 121 sets the threshold value which is used for the filteringprocess on the neighboring pixels of the current block of the intra prediction performed 5 in the filtering processing unit 122 and supplies the threshold value information representing the set threshold value (after-updating threshold value) to the filtering processing unit 122, [0102]
10 For example, the threshold value setting unit 121 may-set the threshold value according to the bit depth of the image data which are the encoding target. [0103]
In this case, for example, the threshold value setting
15 unit 121 may acguire the information on the bit depth of the image data v/'hich are parameters transmitted as the sequence parameter set (SPS) or the liJce from the lossless encoding unit 106 to the decoding side, determine the bit depth of the image data based on the information on the bit depth, and set
20 the threshold value according to the bit depth. In addition, for example, the threshold value setting unit 121 may acquire the image information (image data or information on the image data) from the screen rearrangement buffer 102, determine the bit depth of the image data based on the image information
25 (by analyzing the image information) , and set the threshold value according to the bit depth. [0104]
In addition, in this case, for example, the threshold value setting unit 121 may update the threshold value by
30 bit-shifting the initial value (for example, 8) of a
predetermined threshold value according to the determined bit

1

SP351032WO001
depth. At this time, for example, a value appropriate for the case where the bit depth is 8 bits may be predetermined as the initial value in advance, and the threshold value setting unit 121 may bit-shift the initial value according to a 5 difference in the number of bits betv/een an actual bit depth of the image data and the 8 bits. [0105]
In addition, in this case, for example, the threshold value setting unit 121 may supply the threshold value
10 informatxon representing the after-updating threshold value to the lossless encoding unit 106, so that the sequence parameter set (SPS), the picture parameter set (PPS) or the like may be transmitted to the decoding side. At this time, for example, the threshold value setting unit 121 may encode
15 (for example, golomb-encode) the threshold value information and supply as the threshold value encoding information. In addition, for example, in the case v/here the threshold value setting unit 121 determines the bit depth of the image data based on the image information, and the threshold value setting
20 unit supplies the information on the bit depth to the lossless encoding unit 106, so that the sequence parameter set (SPS), the picture parameter set (PPS), or the like may be transmitted to the decoding side . At this time, for example, the threshold value setting unit 121 may encode (for example, golomb-encode)
25 the information on the bit depth and supply as the bit depth encoding information. [0106]
In addition, in this case, for example, the threshold value setting unit 121 may generate flag information (threshold
30 value change flag) representing v/hether or not the threshold value is updated (changed) and supply a threshold value change

/

SP351032WO001
flag to the lossless encoding unit 106 to transmit the threshold value change flag to the decoding side. By doing sO/ in the decoding side (for example, the image decoding device), it is possible to easily identify based on the value of the 5 threshold value change flag v;hether or not the threshold value is updated (changed) . Namely, in the decoding side (for example, the image decoding device) , it is possible to easily control whether or not to perform the process of updating (changing) the threshold value similarly to the encoding side
10 (for example, the image encoding device 100). [0107]
In addition, for example, the threshold value setting unit 121 may set the threshold value according to external designation such as user's designation. In this case, the
15 value designated by the user or the like corresponds to the above-described after-updating value . Namely, the threshold value corresponding to the value is supplied to the filtering processing unit 122. [0108]
20 In this case, for example, the threshold value setting unit 121 may supply the threshold value information representing the set threshold value to the lossless encoding unit 106, so that the sequence parameter set (SPS) , the picture parameter set (PPS), or the like may be transmitted to the
25 decoding side. At this time, for example, the threshold value setting unit 121 may encode (for example, golomb-encode) the threshold value information and supply as the threshold value encoding information. [0109]
30 In addition, in this case, for example, the threshold value setting unit 121 may generate a threshold value change

1

n.i
3^
SP351032WO001

flag and supply the threshold value change flag to the lossless encoding unit 106 to transmit the threshold value change flag to the decoding side. [0110] 5 In addition, for example, the threshold value setting unit 121 may update (change) the threshold value externally designated by a user or the like according to the bit depth of the image data v/hich is the encoding target. [0111]
10 In this case, for example, the threshold value setting unit 121 may acquire the information on the bit depth of the image data from the lossless encoding unit 106 and determine the bit depth of the image data based on the information on the bit depth. In addition, for example, the threshold value
15 setting unit 121 may acquire the image information from the screen rearrangement buffer 102 and determine the bit depth of the image data based on the image information. [0112]
In addition, in this case, for example, the threshold
20 value setting unit 121 may update the threshold value by bit-shifting the threshold value externally designated by a user or the like according to the determined bit depth. At this time, for example, the threshold valii.e appropriate to the case v/here the bit depth is 8 bits may be designated, and
25 the threshold value setting unit 121 may bit-shift the
designated threshold value according to a difference in the number of bits betv/een an actual bit depth of the image data and 8 bits. [0113]
30 For example, the threshold value externally designated by a user or the like is defined as

-3^ %0r
SP351032WO001
contouring_artefact_threshold. The
contouring_artefact_threshold is designated as the value corresponding to the case v/here the bit depth of the image data v/hich is the encoding target is 8 bits. In the case v/here 5 the actual bit depth of the image data is n bits {n > 8) , the threshold value setting unit 121 bit-shifts the contouring_artefact_threshold by using the f ollov/ing Formula {10} . [0114]
10 [Mathematical Formula 5]
contouring_artefact_threshold << {n - 8) ... (10) [0115]
In addition, in this case, for example, the threshold value setting unit 121 may supply the threshold value
15 information representing the after-updating threshold value to the lossless encoding unit 106, so that the sequence parameter set (SPS), the picture parameter set {PPS}, or the like may be transmitted to the decoding side. At this time, for example, the threshold value setting unit 121 may encode
20 (for example, golomb-encode) the threshold value information and supply as the threshold value encoding information. [0116]
In addition, in this case, for example, thft threshold value setting unit 121 may supply the threshold value
25 information representing the before-updating threshold value (the threshold value designated by an external side such as a user) to the lossless encoding unit 106, so that the sequence parameter set (SPS), the picture parameter set {PPS}, or the like may be transmitted to the decoding side. At this time,
30 for example, the threshold value setting unit 121 may encode {for example, golomb-encode) the threshold value information

T

^

^3

SP351032WO001

and supply as the threshold value encoding information. [0117]
At this time/ in addition/ the threshold value setting unit 121 may supply the information on the bit depth to the 5 lossless encoding unit 106/ so that the sequence parameter set (SPS)/ the picture parameter set (PPS)/ or the like may be transmitted to the decoding side . At this time/ for example/ the threshold value setting unit 121 may encode (for example/ golomb-encode) the information on the bit depth and supply
10 as the bit depth encoding information. [0118]
In addition/ in this case/ for example/ the threshold value setting unit 121 generates the threshold value change flag and supplies the threshold value change flag to the
15 lossless encoding unit 106/ so that the sequence parameter set (SPS)/ the picture parameter set (PPS)/ or the lilce may be transmitted to the decoding side. [0119]
In addition/ the threshold value setting unit 121 may
20 perform the setting (updating) of the threshold value based on arbitrary parameters other than the bit depth. In addition/ although the according to initial value of the threshold value is arbitrary/ the initial value may be/ for example,- "8". In addition/ by setting "0" as the threshold value/ the threshold
25 value setting unit 121 prohibits the bi-linear interpolation process from being applied/ so that the filtering process described v/ith reference to Fig. 5 can be applied. Namely/ in the case v/here the threshold value is "0"/ the method disclosed in Non-Patent Document 2 is disabled.
30 [0120]
The filtering processing unit 122 acquires the threshold

SP351032WO001
value information from the threshold value setting unit 121 and performs the filtering process on the neighboring pixels of the current block v/hich is a target of the intra prediction process by using the threshold value. For example, the 5 filtering processing unit 122 may perform the threshold value determining process expressed by Formulas (3) and (4) by using the threshold value acquired from the threshold value setting unit 121 to identify features of the neighboring pixels of the current block.
10 [0121]
In addition, for example, in the case v/here the determination result is false, that is, in the case V7here the periphery of the current block is determined not to be a portion v/here the change of luminosity, color, density, or the like
15 is planar, the filtering processing unit 122 may perform the filtering process (sometimes, referred to a lov/pass filtering process) described v/ith reference to Fig. 5 on the neighboring pixels. [0122]
20 In addition, for example, in the case where the
determination result is true, that is, in the case v/here the periphery of the current block is determined to be a portion v/here luminosity, color, density, and the like are planar, the filtering processing unit 122 may perform a bi-linear
25 interpolation process {sometimes, referred to as a bi-linear filtering process) as expressed in Formulas (5) to (9) instead of the lov/ pass filtering process. [0123]
Once the filtering processing unit 122 acquires the
30 neighboring pixels of the current block v/hich is a target of the process from the intra prediction unit 113, the filtering

SP351032WO001
processing unit performs the above-described filtering process on the neighboring pixels. Next, the filtering processing unit 122 supplies the after-flitering-process neighboring pixels to the intra prediction unit 113 . The intra 5 prediction unit 113 performs intra prediction by using the after-filter ing-process neighboring pixels . By doing so, the intra prediction unit 113 may generate a predicted image reflecting a result of the filtering process. [0124]
10 Namely, as described above, since the threshold value setting unit 121 can set the threshold value for identifying the features of the neighboring pixels of the current block in the intra prediction process at the time of encoding the image data according to the bit depth of the image data or
15 the like, the intra prediction unit 113 can generate the predicted image reflecting the result of the filtering process appropriate to the image data. Namely, the image encoding device 100 can suppress occurrence of noise such as bending in the decoded image and can suppress a deterioration in image
20 quality of the decoded image. [0125]
In addition, as described above, since the threshold value setting unit 121 can set the threshold value according to external designation such as user' s designation, the intra
25 prediction unit 113 can reflect the external designation such as the user ' s designation of the image quality on the predicted image. Namely, the image encoding device 100 can control the image quality of the decoded image. [0126]
30 In addition, as described v;ith reference to Fig. 5, Formulas (5) to (9) , and the like, themethodv/here the filtering

¥

SP351032WO001
processing unit 322 applies the filtering process to the neighboring pixels of the current block may be controlled according to the intra prediction mode (namely, the block size of the current block). 5 [0127]
In addition, in the case v/here the bit depth of the image data is as small as, for example, 8 bits, the bending (contour distortion) illustrated in Fig. 7 is remarkably observed. Hov^ever, in the case v;here the bit depth is as large as, for
10 example, 10 bits, the bending is suppressed (is not visually conspicuous) . Therefore, an upper limit of the bit depth v/ith which the bi-linear filtering process disclosed in Non-Patent Document 2 is applied may be provided. For example, the bi"linear filtering process disclosed in Non-Patent Document
15 2 may be applied to only the case v;here the bit depth is 8 bits, and the bi-linear filtering process may not be applied to the other cases. [0128]
In addition, the bi-linear filtering process disclosed
20 in Non-Patent Document 2 may be applied to only the processing of brightness signals. Hov/ever, the bi-linear filtering process may be applied to the process of color difference signals . Namely, the present techniquemay be applied to color difference signals as v/ell as brightness signals.
25 [0129]
In addition, in the case v/here the input signal is 4:4:4 or RGB and each color component channel is independently processed, the present technique may be independently applied to each channel.
30 [0130]
In addition, in the case of per forming hierarchical image

SP351032WO001
encoding (scalable encoding)/hierarchical image decoding (scalable decoding), for example, a threshold value, parameters such as a bit depth, a flag and the like may be allov;ed to be transmitted to only the base layer, and in the 5 non-base layer (enhancement layer) , the parameters, the flag, and the like transmitted to the base layer may be allov;ed to be referred to. [0131]

10 H'ig. 11 is a block diagram illustrating a main
configuration example of the threshold value setting unit 121 and the filtering processing unit 122 in the case v/here the threshold value setting unit 121 of Fig. 10 sets the threshold value according to the bit depth of the image data v/hich is
15 a target of the encoding. [0132]
In the example of Fig. 11, the threshold value setting unit 121 is configured to include a bit depth determining unit 131, a threshold value bit shifting unit 132, and an encoding
20 unit 133. [0133]
The bit depth determining unit 131 determines the bit depth of the image data as the encoding target and supplies the information representing the bit depth to the threshold
25 value bit shifting unit 132. [0134]
For example, the bit depth determining unit 131 acquires the information on the bit depth of the image data from the lossless encoding unit 106 and determines the bit depth of
30 the image data based on the information on the bit depth. In addition, for example, the bit depth determining unit 131

SP351032WO001
acquires image information from the screen rearrangement buffer 102 and determines the bit depth of the image data based on the image information. [0135] 5 In addition, in the case of transmitting the information representing the determined bit depth to the decoding side, the bit depth determining unit 131 also supplies the information representing the bit depth to the encoding unit 133.
10 [0136]
The threshold value bit shifting unit 132 updates (changes) the threshold value by bit-shifting a predetermined threshold value (initial value) v/hich is defined in advance according to the information representing the bit depth
15 supplied from the bit depth determining unit 131. [0137]
For example, in the threshold value bit shifting unit 132, the value appropriate to the case where the bit depth is 8 bits is set as the initial value of the threshold value
20 in advance. Once the threshold value bit shifting unit 132 acquires the information representing the bit depth from the bit depth determining unit 131, the threshold value bit shifting unit bit-shifts the initial value by a difference in the number of bits betv/een the bit depth of the image data
25 represented by the information and the 8 bits. [0138]
The threshold value bit shifting unit 132 supplies the information (threshold value information) representing the after-updating (changing) threshold value to the neighboring
30 pixel determining unit 142 of the filtering processing unit 122.

SP351032WO001
[0139]
In addition, in the case of transmitting the threshold value information representing the after-updating (changing) threshold value to the decoding side, the threshold value bit 5 shifting unit 132 also supplies the threshold value information to the encoding unit 133. [0140]
The encoding unit 133 golomb-encodes the supplied information and supplies the obtained golomb code to the
10 lossless encoding unit 106 to transmit the golomb code to the decoding side. For example, in the case of transmitting the information representing the bit depth to the decoding side, the encoding unit 133 acquires the information representing bit depth from the bit depth determining unit 131 and performs
15 golomb encoding on the information representing the bit depth . The encoding unit 133 supplies the obtained golomb code of the information representing the bit depth (sometimes, referred to as bit depth encoding information) to the lossless encoding unit 106 to transmit the golomb code to the decoding
20 side. [0141]
In addition, for example, in the case of transmitting the threshold value information representing the after-updating (changing) threshold value to thedecoding side,
25 the encoding unit 133 acquires the threshold value information representing the after-updating (changing) threshold value from the threshold value bit shifting unit 132 and golomb-encodes the threshold value information . The encoding unit 133 supplies the obtained golomb code of the threshold
30 value information (sometimes, referred to as threshold value encoding information) to the lossless encoding unit 106 to

SP351032WO001
transmit the golomb code to the decoding side. [0142]
In addition, the encoding unit 133 may generate a threshold value change flag representing v/hether or not the 5 threshold value is updated (changed) and supply the threshold value change flag to the lossless encoding unit 106 to transmit the threshold value change flag to the decoding side. [0143]
In addition, as illustrated in Fig. 11, the filtering 10 processing unit 122 is configured to include a mode/block size buffer 141, a neighboring pixel determining unit 142, a filter decision unit 143, a lov7 pass filter unit 144, and a bx~linear filter unit 145. [0144] 15 The mode/block size buffer 141 acguires the information (mode/block size) on the block size and the mode of the current block V7ith respect to the prediction modes v/hich are candidates from the intra prediction unit 113 and stores the information. [0145] 20 At a predetermined timing or based on an external reguest, the mode/block size buffer 141 supplies the stored information (block size) on the block size to the neighboring pixel determining unit 142 . In addition, at a predetermined timing or based on an external reguest, the mode/block size buffer 25 141 supplies the stored information (mode) on the mode and the stored information (block size) on the block size to the filter decision unit 143. [0146]
The neighboring pixel determining unit 142 acguires the
30 neighboring pixels adjacent to the upper side and the left
side of the current block v/ith respect to the prediction modes

SP351032WO001
which are candidates from the intra prediction unit 113. In addition, the neighboring pixel determining unit 142 acquires the threshold value information from the threshold value bit shifting unit 132. In addition, the neighboring pixel 5 determining unit 142 acguires the information {block size) on the block size from the mode/block size buffer 141. [0147]
In the case of the mode where the current block has a predetermined size (for example, 32 x 32) (or a size within
10 a predetermined range) , the neighboring pixel determining unit 142 performs the threshold value determining process for selecting the filter which is to be used for the filtering process on the neighboring pixels acguired from the intra prediction unit 113 based on the information on the block size
15 acquired from the mode/block size buffer 141 by using the threshold value information acquired from the threshold value bit shifting unit 132. Namely, the neighboring pixel determining unit 142 determines features of the neighboring pixels (for example, determines v;hether or not to be pixels
20 of a portion v/here the change of luminosity, color, density, or the like is planar). [0148]
The neighboring pixel determining unit 142 supplies the determination result to the filter decision unit 143. In
25 addition, the neighboring pixel determining unit 142 supplies the neighboring pixels of the current block acguired from the intra prediction unit 113 to the lov; pass filter unit 144 and the bi-linear filter unit 145. [0149]
30 The filter decision unit 143 acquires the information (mode) on the mode and the information (block size) on the

SP351032WO001
block size from the mode/block size buffer 141. In addition^ the filter decision unit 143 acquires the determination result of the threshold valuedeterminingprocess f romthe neighboring pixel determining unit 142. The filter decision unit 143 5 decides the type of the to-be-executed filtering process and the application method thereof by using the acquired information and result. For example^ the filter decision unit 143 determine whether or not any of the lov/ pass filtering process and the bi-linear filtering process is applied to the
10 neighboring pixels of the current block, how to perform the filtering process^ and the like. [0150]
The filter decision unit 143 supplies the control information v/hich controls execution of the filtering process
15 to the lov; pass filter unit 144 and the bi-linear filter unit 145 according to the decision. Namely^ the filter decision unit 143 supplies control information indicating hov/to perform the filtering process to the processing unit selected betv/een the lov; pass filter unit 144 and the bi-linear filter unit
20 145 and supplies control information indicating to stop the filtering process (that is^. control information indicating not to perform the filtering process) to the non-selected processing unit. [0151]
25 The lov; pass filter unit 144 performs the lov; pass filtering process described v;ith reference to Fig. 5 on the neighboring pixels of the current block supplied from the neighboring pixel determining unit 142 according to the control information supplied from the filter decision unit 143. The
30 lov; pass filter unit 144 supplies the neighboring pixels (after-filtering-process neighboring pixels) v;hich are

SP351032WO001
applied v/ith the lov; pass filtering process to the intra
prediction unit 113.
[0152]
The bi-linear filter unit 145 performs the bi~linear
5 filtering process described v;ith reference to Formulas (5)
to (9) on the neighboring pixels of the current block supplied
from the neighboring pixel determining unit 142 according to
the control information supplied from the filter decision unit
143. The bi-linear filter unit 145 supplies the neighboring
10 pixels (after-filtering-process neighboring pixels) v;hich
are applied vjith the bi-linear filtering process to the intra
prediction unit 113.
[0153]
By doing so, the threshold value setting unit 121 can 15 set the threshold value according to the bit depth of the image data (appropriate to the bit depth) , the filtering processing unit 122 can select the filter appropriate to the bit depth of the image data and perform the filtering process on the neighboring pixels by using the filter, and the intra 20 prediction unit 113 can generate the predicted image by using the neighboring pixels applied with the filtering process appropriate to the bit depth of the image data. Namely, by doing so, the Image encoding device 100 can suppress a deterioration in image quality of the decoded image. 25 [0154]

Next, an example of a flov/ of the encoding process performed by the image encoding device 100 illustrated in Fig. 10 v/ill be described v/ith reference to a flov/chart of Fig. 30 12.
[0155]

SP351032WO001
Instep SlOl, the A/D converter 101 A/D-converts an input image. In step S102, the screen rearrangement buffer 102 stores the A/D-converted image and performs rearrangement from the display order of each picture to the encoding order. 5 [0156]
In step S103, the threshold value setting unit 121 sets the threshold value for identifying the features of the neighboring pixels of the current block in the intra prediction process at the time of encoding the image data.■
10 [0157]
In step S104, the intra prediction unit 113 performs the intra prediction process in the intra prediction mode. Here, in some cases, the filtering processing unit 122 performs the filtering process on the neighboring pixels of the current
15 block by using the threshold value set in step S103. In the case v/here the filtering process is performed, the intra prediction unit 113 performs the intra prediction by using the neighboring pixels v/hich are applied v;ith the filtering process.
20 [0158]
In step S105, the inter prediction unit 114 performs the inter prediction process for performing the motion prediction or the motion compensation in the inter prediction mode.
25 [0159]
In step S106, the predicted image selection unit 115 decides the optimal mode based on the cost function values output from the intra prediction unit 113 and the inter prediction unit 114 . Namely, the predicted image selection
30 unit 115 selects any one of the predicted image generated by the intra prediction unit 113 and the predicted image generated

SP351032WO001
by the inter prediction unit 114. [0,160]
In step S107/ the arithmetic unit 103 calculates a difference betv/een the image rearranged by the process of step 5 S102 and the predicted image selected by the process of step S106. The data amount of the difference data is decreased in comparison with the data amount of the original image data. Therefore, in comparison vjith the case v?here the image is encoded in the state v/here the image is not processed, the
10 data amount can be compressed. [0161]
In step S108, the orthogonal transform unit 104 performs orthogonal transform on the difference information generated by the process of step S106. In step S109, the quantization
15 unit 105 performs quantization on the orthogonal transform coefficient obtained by the process of step S108 by using the quantization parameter calculated by the rate control unit 116. [0162]
20 In step SllO, the inverse quantization unit 108 performs inverse quantization on the quantized coefficient (sometimes, referred to as quantization coefficient) generated by the process of step S109 by using characteri-sti cs corresponding to the characteristics of the quantization unit 105. In step
25 Sill, the inverse orthogonal transform unit 109 performs inverse orthogonal transform on the orthogonal transform coefficient obtained by the process of step SllO. [0163]
In step S112, the arithmetic unit 110 adds the predicted
30 image to locally decoded difference information to generate a locally decoded image {image corresponding to an input to

t-^

SP351032WO001
arithmetic unit 103}.
[0164J
In step S113, the loop filter 111 performs a filtering
process on the image generated by the process of step S112 . 5 Accordingly, the block distortion or the like is removed.
[0165]
In step Sll4, the frame memory 112 stores the decoded
image v/hich is applied with a loop filtering process by the
process of step S113. 10 [0166]
In step S115, the lossless encoding unit 106 encodes
the coefficient quantized by the process of step SI 09 . Namely,
lossless encoding such as variable length encoding or
arithmetic encoding is performed on the data corresponding 15 to the difference image.
[0167]
In addition, at this time, the lossless encoding unit
106 encodes the information on the prediction mode of the
predicted image selected by the process of step S106 and adds 20 the information to the encoded data obtained by encoding the
difference image. Namely, the lossless encoding unit 106
encodes the optimal intra prediction mode information supplied
from the intra prediction unit 113 or encodes the information
according to the optical inter prediction mode supplied from 25 the inter prediction unit 114 to add the formation to the encoded
data.
[0168]
In step S116, the accumulation buffer 107 accumulates
the encoded data obtained by the process of step S115. The 30 encoded data accumulated in the accumulation buffer 107 are
appropriately read out and transmitted through a transmission

1

SP351032WO001
line or a recording medium to the decoding side. [0169]
In step S117, the rate control unit 116 controls the rate of quantization operations of the quantization unit 105 5 based on the code amount {occurring code amount) of the encoded data accumulated in the accumulation buffer 107 by the process of step S116, so that overflovj or underflov/ does not occur. In addition, the rate control unit 116 supplies the information representing the quantization parameter to the quantization
10 unit 105. [0170]
Once the process of step S117 is ended, the encoding process is ended. [0171]
15
Next, an example of a f lovj of the threshold value setting process performed in step S103 of Fig. 12 in an example v/here the threshold value setting unit 121 of the example of Fig. 11 determines the bit depth of the image data based on the
20 information on the bit depth will be described v/ith reference to a flov/chart of Fig. 13, In addition, in Fig. 13, the case v/'here the threshold value encoding information representing the after-updating threshold value and the threshold value change flag are transmitted to the decoding side v/ill be
25 described. [0172]
Once the threshold value setting process is started, in step S131, the threshold value bit shifting unit 132 determines whether or not a predetermined initial value is
30 set as the threshold value. In the case v;here it is determined that the threshold value is updated, the procedure proceeds

to step S132.
[0173]
In step S132, the bit depth determining unit 131 acquires information on the bit depth from the lossless encoding unit 106.
[0174]
In step 3133/ the bit depth determining unit 131 determines the bit depth of the image data v;hich is a target of the encoding based on the information on the bit depth acguired in step S132.
[0175]
In step S134, the threshold value bit shifting unit 132 bit-shifts the threshold value (initial value} according to the determination result {determined bit depth) of stepS133. The threshold value bit shifting unit 132 supplies the after-updating threshold value to the neighboring pixel determining unit 142 of the filtering processing unit 122.
[0176]
In step S135, the encoding unit 133 encodes the threshold value information representing the after-bit-shifting
(after-updating) threshold value generated in step S134.
[0177]
In step 3136/ the encoding unit 133 supplies the threshold value encoding information representing the after-updating threshold value obtained by the process of step 3135 to the lossless encoding unit 106 to transmit the threshold value encoding information to the decoding side.
[0178]
Once the process of step 3136 is ended, the procedure proceeds to step 3137. In addition, in the case v/here it is determined in step 3131 that the initial value is set as the

threshold value, the procedure proceeds to step S137.
[0179]
In step S137, the encoding unit 133 sets the threshold value change flag representing as to v/hether or not the threshold value is updated (changed). Namely, for example, in the case of performing the processes of steps S131 to S136, the encoding unit 133 sets the threshold value change flag of the value representing that the threshold value is updated
(changed) . In addition, for example, in the case v/here it is determined in step S131 that the initial value is set as the threshold value, the encoding unit 133 sets the threshold value change flag of the value representing that the threshold value is not updated (changed}.
[0180]
In step S138, the encoding unit 133 supplies the threshold value change flag set in step S137 to the lossless encoding unit 106 to transmit the threshold value change flag to the decoding side.
[0181]
Once the process of step S138 is ended, the threshold value setting process is ended, and the procedure returns to Fig. 12.
[0182]
By performing the threshold value setting process as described above, the threshold value setting unit 121 can set the threshold value according to the bit depth of the image data (appropriate to the bit depth).
[0183]
In addition, in the case v/here the threshold value change flag is not transmitted to the decoding side (that is, in the case v/here the threshold value change flag is not generated) ,

the above-described processes of steps S137 and S138 may be
omitted.
[0184]
In addition, in the case where the threshold value encoding information representing the after-updating threshold value is not transmitted to the decoding side (that is, in the case v/here the threshold value encoding information is not generated) , the above-described processes of steps S135 and S136 may be omitted. [0185]

Next, an example of a f lev/of the intra prediction process performed in step S104 of Fig. 12 v/ill be described v/ith reference to a flov/chart of Fig. 14. [0186]
Once the intra prediction process is started, in step S151, the neighboring pixel determining unit 142 determines the features of the neighboring pixels of the current bloc]<; acquired from the intra prediction unit 113 (for example, features as to v/hether or not the periphery of the current block is a portion v;here luminosity, color, density, or the like is planar) based on the block size of the mode/block size buffer 141 and the threshold value information acquired from the threshold value bit shifting unit 132. [0187]
In step S152, the filter decision unit 143 determines the filter v/hich is to be used for the neighboring pixels of the current block based on the mode and block size acquired from the mode/block size buffer 141 and the determination result of step S151. [0188]

In step S153, the filter unit selected among the lev/ pass filter unit 144 and the bi-linear filter unit 145 by the process of step S152 performs the filtering process on the neighboring pixels of the current block. For example, in the case where the lov/ pass filter is selected in step S152, in step S153, the lov? pass filter unit 144 performs the low pass filtering process on the neighboring pixels in the method decided by the filter decision unit 143. In addition, for example, in the case where the bi-linear filter is selected in step S152, in step S153, the bi-linear filter unit 145 performs the bi-linear filtering process on the neighboring pixels in the method decided by the filter decision unit 143 . [0189]
In addition, the filtering process may be omitted in some modes. [0190]
In step S154, the intra prediction unit 113 performs the intra prediction in the modes by using the neighboring pixels applied with the filtering process in step S153 (in the case v/here the filtering process is omitted, the neighboring pixels which is not applied with the filtering process) . [0191]
In step S155, the intra prediction unit 113 calculates the cost function values v/ith respect to the results of the intra prediction of the modes performed in step S154. [0192]
In step S156, the intra prediction unit 113 determines the optimal intra prediction mode based on the cost function values of the modes calculated in step S155. [0193]

In step S157, the intra prediction unit 113 generates the predicted image in the optimal intra prediction mode determined in step S15 6. [0194]
Once the process of step S157 is ended, the intra prediction process is ended, and the procedure returns to Fig. 12. [0195]
By performing the above-described intra prediction process, the filtering processing unit 122 can select the filter appropriate to the bit depth of the image data andperform the filtering process on the neighboring pixels by using the filter, and the intra prediction unit 113 can generate the predicted image by using the neighboring pixels vjhich are applied v/ith the filtering process appropriate to the bit depth of the image data. [0196]
Namely, by performing the above-described processes, the image encoding device 100 can suppress a deterioration in image quality of the decoded image. [0197]

Next, an example of a f lov7 of the threshold value setting process performed in step S103 of Fig. 12 in an example where the threshold value setting unit 121 of the example of Fig. 11 determines the bit depth of the image data based on the image information v/ill be described with reference to a flowchart of Fig. 15. In addition, in Fig. 15, the case v;here the threshold value encoding information representing the after-updating threshold value and the threshold value change flag are transmitted to the decoding side vjill be described.

[0198]
In the case of the example of Fig. 15, processes of steps S171, S174 to S178 are performed similarly to the processes of steps S131, S134 to S138 of the example of Fig. 13. [0199]
In the case v/here it is determined in step S171 of Fig. 15 that the threshold value is updated, the procedure proceeds to step S172. [0200]
In step S172, the bit depth determining unit 131 acquires the image information from the screen rearrangement buffer 102. [0201]
In step 3173/ the bit depth determining unit 131 determines the bit depth of the image data v/hich is the encoding target based on the image information acquired in step S172 . Once the process of step S173 is ended, the procedure proceeds to step S174 . [0202]
Once the process of step S178 is ended, the threshold value setting process is ended, and the procedure returns to Fig. 12. [0203]
By doing so, the threshold value setting unit 121 can also set the threshold value according to the bit depth of the image data (appropriate to the bit depth). Therefore, in this case, the image encoding device 100 can also suppress a deterioration in image quality of the decoded image. [0204]
In addition, in the case v/here the threshold value change flag is not transmitted to the decoding side (namely, the

-^^

SP351032WO001

threshold value change flag is not generated) , the processes
of steps 3177 and S178 may be omitted.
[0205]
In addition, in the case v/here the threshold value 5 encoding information representing the after-updating
threshold value is not transmitted to the decoding side (namely, the threshold value encoding information is not generated), the processes of steps S175 and S176 may be omitted. [0206] 10
Next, an example of a f lov; of the threshold value setting process performed in step S103 of Fig. 12 in an example v^here the threshold value setting unit 121 of the example of Fig. 11 determines the bit depth of the image data based on the 15 image information and transmits the determined bit depth to the decoding side will be described with reference to a flov/chart of Fig. 16. In addition, in Fig. 16, the case v/here the threshold value change flag is also transmitted to the decoding side will be described. 20 [0207]
In the case of the example of Fig. 16, processes of steps S191 to S194, S197, and S198 are performed similarly to the processes of steps S171 to S174, S177, and 3178 of the example of Fig. 15. 25 [0208]
Once the process of step 3194 of Fig. 16 is ended, the procedure proceeds to step 3195. [0209J
In step 3195, the encoding unit 133 encodes the 30 information representing the bit depth determined in step 3193 . [0210]

In step 3196/ the encoding unit 133 supplies the bit depth encoding information obtained by the process of step S195 to the lossless encoding unit 106 to transmit the bit depth encoding information to the decoding side. [0211]
Once the process of step 3196 is ended/ the procedure proceeds to step 3197. [0212]
By doing sO/ the threshold value setting unit 121 can also set the threshold value according to the bit depth of the image data {appropriate to the bit depth). Therefore/ in this case, the image encoding device 100 can also suppress a deterioration in image quality of the decoded image. [0213]
In addition/ in the case v/here the threshold value change flag is not transmitted to the decoding side (namely/ the threshold value change flag is not generated) / the processes of steps 3197 and 3198 may be omitted. [0214]
In addition/ in the case where the bit depth encoding information is not transmitted to the decoding side (namely/ the bit depth encoding information is not generated)/ the processes of steps S195 and 3196 may be omitted. [0215]

Fig. 17 is a block diagram illustrating a main configuration example of the threshold value setting unit 121 and the filtering processing unit 122 in the case v/here the threshold value setting unit 121 of Fig. 10 sets the threshold value according to external designation such as user's designation.

[0216]
In the example of Fig. 17^ the threshold value setting unit 121 is configured to include a threshold value setting unit 151 and an encoding unit 133. [0217] .
The threshold value setting unit 151 receives external designation of the threshold value, for example, user's designation and sets the threshold value according to the designation. The threshold value setting unit 151 supplies the information {threshold value information) representing the set threshold value to the neighboring pixel determining unit 142 of the filtering processing unit 122. [0218]
In addition, in the case of transmitting the threshold value information to the decoding side, the threshold value bit shifting unit 132 also supplies the threshold value information to the encoding unit 133. [0219]
Similarly to the case of Fig. 11, the encoding unit 133 golomb-encodes the supplied information and supplies the obtained golomb code to the lossless encoding unit 106 to transmit the golomb code to the decoding side. For example, in the case of transmitting the threshold value information representing the set threshold value to the decoding side, the encoding unit 133 acguires the supplied threshold value information from the threshold value setting unit 151 and golomb-encodes the threshold value information. The encoding unit 133 supplies the obtained threshold value encoding information to the lossless encoding unit 106 to transmit the threshold value information to the decoding side. [0220]

In addition, similarly to the case of Fig. 11, the encoding unit 133 may generate a threshold value change flag representing v/hether or not the threshold value is updated (changed) and supply the threshold value change flag to the lossless encoding unit 106 to transmit the threshold value change fl^g to the decoding side.
[0221]
In addition, the filtering processing unit 122 has the same configuration as that of the case of Fig. 11 and performs the same processes as those of the case of Fig. 11.
[0222]
By doing so, the threshold value setting unit 121 can set the threshold value according to the bit depth of the image data (appropriate to the bit depth) , the filtering processing unit 122 can select the filter appropriate to the bit depth of the image data and perform the filtering process on the neighboring pixels by using the filter, and the intra prediction unit 113 can generate the predicted image by using the neighboring pixels v^hich are applied vjith the filtering process appropriate to the bit depth of the image data . Namely, by doing ^o, image encoding device 100 can suppress a deterioration in image quality of the decoded image. [0223]
In addition, in the case of the example of Fig. 17, since the threshold value setting unit 121 can set the threshold value according to external designation such as user's designation, the intra prediction unit 113 can reflect external designation such as user's designation of the image quality on the predicted image. Namely, the image encoding device 100 can control the image quality of the decoded image.

CLAIMS
1. An image processing apparatus comprising:
a threshold value setting unit v/hich sets a threshold
value ivhich is compared v/ith a value calculated by using a sample of neighboring pixels in order to identify a feature of the neighboring pixels of a current block in an intra prediction process in encoding of image data according to a bit depth of the image data; and
a filtering process unit which performs a filtering process on the neighboring pixels by using a bi-linear interpolation filter in the case v;here the value calculated by using the sample of the neighboring pixels is lovier than the threshold value set by the threshold value setting unit.
2. The image processing apparatus according to claim 1, wherein the threshold value setting unit bit-shifts the threshold value which is determined as an initial value in advance according to the bit depth.
3. The image processing apparatus according to claim 2, wherein the threshold value setting unit sets the threshold value to 8 in the case v/here the bit depth of the image data is 8 bits.
4. The image processing apparatus according to claim 1, further comprising a transmitting unit which transmits the threshold value set by the threshold value setting unit.
5. The image processing apparatus according to claim 1, further comprising a determining unit which determines the

bit depth of the image data,
v/herein the threshold value setting unit sets the threshold value according to the bit depth determined by the determining unit.
6. The image processing apparatus according to claim 5,
further comprising a transmitting unit v/hich transmits the
bit depth determined by the determining unit.
7. The image processing apparatus according to
claim 1, v/herein the filtering process unit performs a low
pass filtering process on the neighboring pixels in the case
where the value calculated by using the sample of the
neighboring pixels exceeds the threshold value set by the
threshold value setting unit.
($• hn image processing method comprising:
setting a threshold value v/hich is compared v/ith a value calculated by using a sample of neighboring pixels in order to identify a featur^e of the neighboring pixels of a current block in an intra prediction process in encoding of image data according to a bit depth of the image data; and
.performing a filtering process on the neighboring pixels by using a bi-linear interpolation filtez~ in the case v/here the value calculated by using the sample of the neighboring pixels is lovfer than the set threshold value.