Description
Title of Invention: IMAGE PROCESSING DEVICE,
PROJECTOR, AND METHOD OF CONTROLLING
PROJECTOR
Technical Field
[0001] The present invention relates to a projector for displaying an image by projecting the
image on a projection surface.
Background Art
[0002] In the case of displaying an image by projecting the image on a projection surface
such as a screen using a projector, an adjustment corresponding to a relative positional
relationship between the projector and the projection surface is generally performed.
As such an adjustment, there can be cited a focus adjustment for adjusting the shift of
the focus of the image (hereinafter also referred to as a "projection image") projected
on the projection surface, a keystone correction for correcting the distortion
(hereinafter also referred to as a "keystone distortion") of the image range of the
projection image, and so on.
[0003] The focus adjustment and the keystone correction described above are performed by
taking an image of a test pattern projected on the projection surface with a camera
installed in the projector, and then obtaining the information necessary for the focus
adjustment and the keystone correction based on the image (hereinafter also referred to
as a "taken image") thus taken (see PTL 1 and PTL 2).
[0004] In PTL 1, the operation is as simply explained below. Specifically, the light with
even luminance is projected from the projector toward the screen, the reflected light
from the screen is received, and the centroid position of the illuminance of the received
light is detected. Then, the tilt angle of the screen is calculated based on the centroid
position thus detected, and then the keystone distortion is corrected in accordance with
the tilt angle thus calculated.
[0005] In PTL 2, the operation is as simply explained below. Specifically, focusing attention
to four sides (upper, lower, right, and left sides) of the screen in the image taken by the
imaging section, the ratios between the lengths of the two sides opposed to each other,
namely the upper and lower sides and the right and left sides, are calculated. The ratio
between the intensities of the lights respectively projected on the upper and lower areas
is calculated based on the ratio between the upper and lower sides, and similarly, the
ratio between the intensities of the lights respectively projected on the right and left
areas is calculated based on the ratio between the right and left sides. Then, the test
pattern based on the ratios of the light intensities is projected.
Citation List
Patent Literature
[0006] PTL 1: JP-A-2005- 159426
PTL 2: JP-A-201 1-176629
Summary of Invention
Technical Problem
[0007] However, in PTL 1 mentioned above, there arise a problem that the detection
accuracy of the reflected light might drop depending on the positional relationship
between the projector and the screen, a problem that the detection accuracy of the
reflected light is significantly different between the locations, and so on. For example,
in the case in which the screen is tilted, since the light intensity decreases in inverse
proportion to the square of the distance from the light source, the longer the distance
(hereinafter also referred to as a "projection distance") of the light projected by the
projector to the screen becomes, the more saliently the drop of the illuminance of the
reflected light received from the screen grows, and the more dramatic drop of the
accuracy of the reflected light received results. Therefore, the sharper the tilt becomes,
the more significant the drop of the illuminance of the reflected light received corre
sponding to the light with longer projection distance is, and the accuracy of the
reflected light received drops to cause the calculation accuracy of the centroid position
to drop as a result. Further, in some cases, the angle of the projection with respect to
the screen is different even if the projection distance is the same, and there is a pos
sibility of failing to detect the difference accurately.
[0008] Further, PTL 2 mentioned above is premised on the concept that the light intensity
variation between the two sides opposed to each other is a gradational variation based
on the ratio between the lengths of the sides. However, according to the verification
with an actual measurement test, it is confirmed that the gradational variation might
fail to occur in some cases. It is possible that generation of the light intensity variation
due to the contamination in the projection surface of the screen, the distortion of the
screen, and so on is the factor thereof. Further, if the gradational variation based on the
ratio between the lengths of the sides fails to occur as described above, there arises a
problem that degradation of the detection accuracy of the test pattern is incurred.
Further, PTL 2 is premised on the fact that the imaging section is integrated with the
projector, and in the case of the separate imaging section not installed in the projector,
the error between the positional relationship between the projector and the screen and
the positional relationship between the imaging section and the screen is accumulated,
and there arises a problem that the possibility of incurring the degradation of the
detection accuracy of the test pattern rises.
Solution to Problem
[0009] An advantage of some aspects of the invention is to provide a technology of further
improving the detection accuracy of the detection image projected on the projection
surface compared to the related art.
[0010] The invention can be implemented as the following forms or application examples.
[0011] Application Example 1
This application example is directed to an image processing device used for a
projector adapted to display an image by projecting the image on a projection surface,
and including a detection image generation section adapted to generate a detection
image, which is an image adapted to detect a state of a projection image displayed on
the projection surface, and includes a plurality of detection image parts, each of the
detection image parts includes a plurality of regions having respective luminance
values different from each other, and the detection image generation section changes a
luminance distribution of each of the detection image parts of the detection image to be
generated so that maximum luminance values of the detection image parts included in
a taken detection image obtained by imaging fall within an allowable range.
In this image processing device, since the detection image part is formed as an image
including a plurality of regions having respective luminance values different from each
other, and luminance distribution of each of the detection image parts of the detection
image to be generated is changed so that the maximum luminance values of the
detection image parts included in the taken detection image fall within the allowable
range, it is possible to perform the adjustment so that the luminance values of the
detection image parts included in the taken detection image become roughly equal to
each other. Thus, it is possible to improve the detection accuracy of the centroid co
ordinate of each of the detection image parts included in the taken detection image, and
it becomes possible to improve the detection accuracy of the detection image.
[0012] Application Example 2
This application example is directed to the image processing device according to Ap
plication Example 1, wherein the detection image generation section takes the
luminance value of either one of the taken detection image parts out of the detection
image parts included in the taken detection image as a reference luminance value, and
changes the luminance distribution of each of the detection image parts of the detection
image to be generated so that the maximum luminance values of the detection image
parts included in the taken detection image are roughly equal to each other based on a
relationship between the reference luminance value and luminance values of other
detection image parts.
In this image processing device, the detection image generation section is capable of
changing the luminance distribution of each of the detection image parts of the
detection image to be generated so that the luminance values of the detection image
parts included in the taken detection image obtained by imaging fall within the
allowable range. Thus, it is possible to improve the detection accuracy of the centroid
coordinate of each of the detection image parts included in the taken detection image,
and it becomes possible to improve the detection accuracy of the detection image.
[0013] Application Example 3
This application example is directed to the image processing device according to Ap
plication Example 1 or 2, wherein the detection image generation section changes the
luminance distribution of each of the detection image parts of the detection image to be
generated by changing at least one of an overall size of the detection image part
included in the detection image, a width of each of the regions included in the
detection image part, and the maximum luminance value of the detection image part.
In this image processing device, by changing at least one of the overall size of the
detection image part, the width of each of the regions included in the detection image
part, and the maximum luminance value of the detection image part, the luminance dis
tribution of each of the detection image parts of the detection image to be generated
can be changed. Thus, it is possible to improve the detection accuracy of the centroid
coordinate of each of the detection image parts included in the taken detection image,
and it becomes possible to improve the detection accuracy of the detection image.
[0014] Application Example 4
This application example is directed to the image processing device according to any
one of Application Examples 1 through 3, wherein sectioning of the regions is obtained
based on a predetermined function.
[0015] Application Example 5
This application example is directed to the image processing device according to Ap
plication example 4, wherein the function is a Gaussian distribution function.
In the image processing device according to one of the application examples 4 and 5,
it is possible to easily form the plurality of detection image parts included in the
detection image as the detection image parts with the luminance distribution suitable
for the detection of the centroid coordinate of each of the detection image parts.
[0016] Application Example 6
This application example is directed to the image processing device according to any
one of Application Examples 1 through 5, wherein a projection image adjustment
section adapted to adjust image quality of the projection image is further provided, and
the projection image adjustment section performs at least one of a plurality of image
quality adjustments including a focus adjustment and a keystone distortion correction
of the projection image based on the process using the detection image.
In the image processing device according to the application example 6, the accurate
image quality adjustment can be performed based on the process using the detection
image with improved accuracy.
[0017] Application Example 7
This application example is directed to a projector adapted to display an image by
projecting the image on a projection surface, and including the image processing
device according to any one of Application Examples 1 through 6, an imaging section
adapted to take the detection image projected on the projection surface, and a
projection section adapted to project the image based on image data output from the
image processing device.
[0018] Application Example 8
This application example is directed to a method of controlling a projector adapted to
display an image by projecting the image on a projection surface, and including (a)
generating a detection image, which is an image adapted to detect a state of a
projection image displayed on the projection surface, and includes a plurality of
detection image parts, and (b) projecting the detection image on the projection surface,
each of the detection image parts includes a plurality of regions having respective
luminance values different from each other, and in step (a), a luminance distribution of
each of the detection image parts of the detection image to be generated is changed so
that maximum luminance values of the detection image parts included in a taken
detection image obtained by taking the detection image projected on the projection
surface fall within an allowable range.
In this method of controlling a projector, since the detection image part is formed as
an image including a plurality of regions having respective luminance values different
from each other, and luminance distribution of each of the detection image parts of the
detection image to be generated is changed so that the maximum luminance values of
the detection image parts included in the taken detection image fall within the
allowable range, it is possible to perform the adjustment so that the luminance values
of the detection image parts included in the taken detection image become roughly
equal to each other. Thus, it is possible to improve the detection accuracy of the
centroid coordinate of each of the detection image parts included in the taken detection
image, and it becomes possible to improve the detection accuracy of the detection
image.
[0019] It should be noted that the invention can be put into practice in a variety of aspects
such as an image display device, a projector, a method of controlling a projector, a
computer program for controlling the projector, or a storage medium storing the
computer program.
Brief Description of Drawings
[fig. l]Fig. 1 is a block diagram schematically showing a configuration of a projector as
an embodiment of the invention.
[fig.2]Fig. 2 is a flowchart showing a detection image adjustment process in the em
bodiment.
[fig.3A]Fig. 3 is an explanatory diagram showing the detection image generated.
[fig.3B]Fig. 3B is an explanatory diagrams showing the detection image generated.
[fig.4]Fig. 4 illustrates explanatory diagrams showing an outline of a detection image
adjustment process performed using the detection image generated in the step S10
shown in Fig. 2.
[fig.5]Fig. 5 is a flowchart showing a procedure of generating dot patterns as detection
image parts constituting the detection image in the step S10 shown in Fig. 2.
[fig.6A]Fig. 6A is an explanatory diagram showing a method of obtaining correction
information.
[fig.6B]Fig. 6B is an explanatory diagram showing a method of obtaining correction
information.
[fig.7A]Fig. 7 is an explanatory diagram showing the method of obtaining correction
information.
[fig.7B]Fig. 7B is an explanatory diagram showing the method of obtaining correction
information.
[fig.7C]Fig. 7C is an explanatory diagram showing the method of obtaining correction
information.
[fig.8A]Fig. 8A is an explanatory diagram showing a procedure of obtaining correction
values used for the correction of the detection image.
[fig.8B]Fig. 8B is an explanatory diagram showing a procedure of obtaining correction
values used for the correction of the detection image.
[fig.9A]Fig. 9A is an explanatory diagram showing a procedure of obtaining a centroid
coordinate in the step S60a shown in Fig. 2.
[fig.9B]Fig. 9B is an explanatory diagram showing a procedure of obtaining a centroid
coordinate in the step S60a shown in Fig. 2.
[fig. 10] Fig. 10 is a flowchart showing a procedure of centroid coordinate calculation in
one centroid coordinate calculation target region.
[fig. 1l]Fig. 11 is a table showing evaluation results with the detection image correction
and evaluation results without the detection image correction in a comparative manner
[fig. 12] Fig. 12 illustrates a plurality of graphs, each showing a variation in the centroid
coordinate over a plurality of times of measurement for the results with the detection
image correction and without the detection image correction, arranged in a comparative
manner.
[fig. 13A]Fig. 13A is an explanatory diagram each showing an example of another
detection image using dot patterns.
[fig.l3B]Fig. 13B is an explanatory diagram each showing an example of another
detection image using dot patterns.
[fig. 14] Fig. 14 is an explanatory diagram showing another example of the detection
image.
[fig.l5]Fig. 15 is an explanatory diagram showing a configuration of displaying
images on a projection surface in a superimposed manner using a plurality of
projectors.
Description of Embodiments
[0021] A. Configuration of Projector
Fig. 1 is a block diagram schematically showing a configuration of a projector as an
embodiment of the invention. The projector PJ is provided with an input operation
section 10, a control circuit 20, an image processing operation circuit 30, an image
projection optical system 40 (projection section), an imaging section 50, and a motion
detection section 60.
[0022] The input operation section 10 is composed of, for example, a remote controller, and
buttons and keys provided to the projector PJ, wherein the buttons and the keys are not
shown in the drawing. The input operation section 10 outputs instruction information
corresponding to the operation by the user to the control circuit 20. For example, the
instruction information of starting a detection image adjustment process described later
is output from the user to the control circuit 20.
[0023] The image projection optical system 40 generates image light representing an image,
and then images the image light on a screen (a projection surface) SC to thereby
project the image in an enlarged manner. The image projection optical system 40 is
provided with an illumination optical system 420, a liquid crystal light valve 440, and a
projection optical system 460.
[0024] The illumination optical system 420 is provided with a light source lamp 422 and a
lamp drive section 424. As the light source lamp 422, a variety of types of selfluminescent
elements such as a discharge emission light source lamp including, for
example, a super-high pressure mercury lamp and a metal halide lamp, a laser source, a
light-emitting diode, or an organic electro luminescence (EL) element can be used. The
lamp drive section 424 drives the light source lamp 422 based on the control by the
control circuit 20.
[0025] The liquid crystal light valve 440 is a light modulation device for modulating the
light emitted from the illumination optical system 420 based on the image data. The
liquid crystal light valve 440 is formed of a transmissive liquid crystal panel having a
plurality of pixels arranged in a matrix. By controlling the operation of the liquid
crystal of each pixel based on a drive signal from a light valve drive section 380 of the
image processing operation circuit 30 described later, the liquid crystal light valve 440
converts the illumination light emitted from the illumination optical system 420 into
the image light representing the image. It should be noted that in the present em
bodiment the liquid crystal light valve 440 includes three liquid crystal light valves
(not shown) for three color components of red (R), green (G), and blue (B). It should
be noted that it is also possible to project a monochrome image using a single liquid
crystal light valve.
[0026] The projection optical system 460 images the image light emitted from the liquid
crystal light valve 440 on the screen SC to thereby project the image on the screen SC
in an enlarged manner. The projection optical system 460 is provided with a projection
lens 462, a lens drive section 464, and a state detection section 466. The projection lens
462 has a focusing lens for focus adjustment and a zoom lens for zoom adjustment,
which are not shown in the drawing, and are configured so as to move in the optical
axis direction, and enlarges the image light emitted from the liquid crystal light valve
440 in accordance with the zoom position of the zoom lens, and then images the image
light in accordance with the focus position of the focusing lens to thereby project the
image represented by the image light on the screen SC in an enlarged manner. The lens
drive section 464 changes the position (hereinafter referred to as a "focus position") in
the optical axis direction of the focusing lens based on the control by the control circuit
20. Further, the lens drive section 464 changes the position (hereinafter referred to as a
"zoom position") in the optical axis direction of the zoom lens. The state detection
section 466 detects the focus position of the focusing lens and the zoom position of the
zoom lens. It should be noted that since the projection optical system 460 has a typical
configuration, the graphical description and the explanation of the specific con
figuration will be omitted.
[0027] The image processing operation circuit 30 is provided with an input processing
section 320, an image display processing section 340, an image memory 360, and the
light valve drive section 380. Based on the control by the control circuit 20, the input
processing section 320 performs A/D conversion on the input image signal supplied
from the external device, if necessary, to thereby convert it into a digital image signal,
which can be processed by the image display processing section 340. Based on the
control by the control circuit 20, the image display processing section 340 writes the
image data, which is included in the digital image signal output from the input
processing section 320, into the image memory 360 frame by frame, and then performs
various image processing such as a resolution conversion process or keystone
correction process when reading it out therefrom. Further, the image display
processing section 340 superimposes detection image data representing the detection
image output from a control section 220 on the image data. The light valve drive
section 380 drives the liquid crystal light valve 440 with the digital image signal input
from the image display processing section 340. It should be noted that it is also
possible to arrange that the light valve drive section 380 is provided to the image
projection optical system 40, but not to the image processing operation circuit 30.
[0028] Based on the control by the control circuit 20, the imaging section 50 takes a
projection image, and then outputs the image signal corresponding to the image thus
taken to the control circuit 20. The projection image is an image obtained by super
imposing the detection image TP (composed of four dot patterns DP as four detection
image parts) on the image PP (indicated by hatching) represented by the digital image
signal input from the input processing section 320 to the image display processing
section 340 as the detection image, and is projected on the screen SC in an enlarged
manner. The imaging section 50 is configured using, for example, a CCD camera
provided with charge coupled device (CCD) as an imaging element. It should be noted
that the detection image will be described later.
[0029] The motion detection section 60 detects motions around the projection axis, in a
vertical direction and a horizontal direction, and stoppage of the motions in the
projector PJ. It should be noted that the motion detection section can be configured
using a variety of sensors capable of detecting the motions and the stoppage of the
motions, such as an angular velocity sensor, an acceleration sensor, or a gyro sensor.
[0030] The control circuit 20 is a computer provided with a CPU, a ROM, and a RAM, and
executes a control program to thereby constitute the control section 220 and an in
formation storage section 260. The control section 220 functions as a variety of control
function sections for respectively controlling the image processing operation circuit 30,
the image projection optical system 40, the imaging section 50, and the motion
detection section 60 with the control program thus executed. The information storage
section 260 functions as a variety of storage sections for storing the information for the
respective control operations. As examples of the control function sections of the
control section 220, Fig. 1 shows a detection image adjustment section 230 for
performing a detection image adjustment described later, and a projection image ad
justment section 240 for performing the adjustment of the image quality of the
projection image such as the focus adjustment and the keystone correction (keystone
distortion correction). The detection image adjustment section 230 operates by
executing the corresponding program in response to the instruction of starting the
detection image adjustment input by the user with the input operation section 10.
Further, as examples of the storage sections of the information storage section 260,
Fig. 1 shows a setting information storage section 262 for storing setting information
used for various control operations by the control section 220, a detection image in
formation storage section 264 for storing detection image information described later,
and a taken image information storage section 266 for storing the image data of the
taken image taken by the imaging section 50.
[0031] The detection image adjustment section 230 is provided with a detection image
generation section 232, an imaging control section 234, an image analysis section 236,
and a centroid coordinate detection section 238. The detection image generation
section 232 generates the image data of the detection image used for the focus ad
justment and the keystone correction. The imaging control section 234 controls the
imaging section 50 to take the projection image including the detection image
projected on the screen SC, and then stores the projection image (hereinafter also
referred to as a "taken image") thus taken into the taken image information storage
section 266. The image analysis section 236 analyzes the taken image. It should be
noted that the image analysis section 236 can be disposed in the detection image
generation section 232. As described later, the centroid coordinate detection section
238 detects the centroid coordinate used for the image quality adjustment such as the
focus adjustment and the keystone correction performed by the projection image ad
justment section 240. The detection image adjustment section 230 will further be
described later.
[0032] It should be noted that in the present embodiment the detection image generation
section 232 and the image analysis section 236 correspond to a detection image
generation section according to the invention. Further, the control circuit 20 and the
image processing operation circuit 30 correspond to an image processing device
according to the invention.
[0033] B. Operation of Projector
Operational Explanation of Detection Image Adjustment
Fig. 2 is a flowchart showing a detection image adjustment process in the present em
bodiment. When the detection image adjustment section 230 (Fig. 1) of the control
section 220 starts the detection image adjustment process, the detection image
generation section 232 of the detection image adjustment section 230 generates the
detection image (step S10). The image data (hereinafter also referred to as "detection
image data") of the detection image thus generated is output to the image display
processing section 340 (Fig. 1) of the image processing operation circuit 30.
[0034] Figs. 3A and 3B are explanatory diagrams showing the detection image generated.
As shown in Fig. 3A, the detection image TP is composed of dot patterns DPI through
DP4 as the four detection image parts to be disposed at predetermined positions in the
four corners of an image formation area 440f composed of the plurality of pixels
(liquid crystal pixels), arranged in a matrix, of the liquid crystal light valve 440. It is
assumed that the coordinates (in the horizontal and vertical directions) of the four
vertexes located upper left, upper right, lower left, and lower right of the image
formation area 440f are represented by (0, 0), (xx, 0), (0, yy), and (xx, yy), re
spectively. In this case, it is assumed that the four dot patterns DPI through DP4 have
the respective centers (centroids) disposed at predetermined coordinates (xl, yl), (x2,
y2), (x3, y3), and (x4, y4) in the coordinate system of the image formation area 440f.
[0035] As shown in Fig. 3B, the dot patterns DPI through DP4 are each a circular pattern
with a dot pattern size (the diameter) equal to sd (the unit is, for example, [pixels]),
sectioned into a plurality of levels of regions, wherein the number of gray levels is stp
(stp is an integer equal to or greater than 3), varying outward from the center thereof,
and each have a multilevel luminance distribution in which the luminance sequentially
decreases in a direction from a central region toward an outer region. In the example
shown in Fig. 3B, the luminance distribution has a shape emulating the Gaussian dis
tribution. It should be noted that it is assumed that a number n of each of the regions is
sequentially assigned as 0 through stp-1 in a direction from the center toward the
outside. The number of the region (the center region) with a first level is n=0, and the
luminance value (e.g., an 8 bit gray-scale value) thereof is represented by V0, and the
radius thereof is represented by rO (the unit is, for example, [pixels]). Similarly, the
number of the region with a second level is n=l, and the luminance value thereof is
represented by VI, and the radius thereof is represented by r l [pixels]. Further, the
number of the region with a third level is n=2, and the luminance value thereof is rep
resented by V2, and the radius thereof is represented by r2 [pixels]. Therefore, the
number of the region with an n-th level is n=0 through stp-1, and the luminance value
thereof is represented by Vn, and the radius thereof is represented by rn [pixels]. It
should be noted that in the case in which the number of pixels [pixels] is odd, the size
sd of the dot pattern is expressed as sd=(2rn) in a range of -rn through -i-rn defining the
center as 0. In contrast, in the case in which the number of pixels is even, the size sd of
the dot pattern is expressed as sd=(2rn)-l in a range of -rn through +(rn-l), or a range
of -(rn-1) through -i-rn. It should be noted that a method of generating each of the dot
patterns DPI through DP4 constituting the detection image TP will further be
described in detail.
[0036] Here, the reason of using such dot patterns as shown in Fig. 3B is that the dot
patterns are suitable to accurately determine the centroid coordinate in the process of
obtaining the centroid coordinate of each of the dot patterns. It should be noted that the
process of obtaining the centroid coordinate will be described later.
[0037] After the generation of the detection image in the step S10 described above, the steps
S20 through S60b are executed, and then the process returns to the step S10 to perform
the process in the step S10 and the subsequent steps again. Thus, the detection image
adjustment is performed. Hereinafter, prior to the specific explanation of the process
on and after the step S20, an outline of the detection image adjustment using the
detection image generated in the step S10 will firstly be explained.
[0038] Fig. 4 illustrates explanatory diagrams showing an outline of a detection image ad
justment process performed using the detection image generated in the step S10 shown
in Fig. 2. The detection image TP generated in the step S10 is displayed by projection
on the screen SC as shown in, for example, (A) in Fig. 4 in the step S20 described
later. On this occasion, it is assumed that the detection image TP displayed by
projection is in the state in which the luminance of the dot patterns DP2, DP4 located
at upper right and lower right positions is lower than the luminance of the dot patterns
DPI, DP3 located at the upper left and lower left positions. In this case, imaging of the
projection image in the step S30 described later and correction of the detection image
in the steps S60b, S10 are performed. As a result, as shown in (B) in Fig. 4, in the
detection image TPa displayed by projection due to reprojection of the detection image
after the correction in the step S20, the adjustment is performed so that the luminance
values of the respective detection image parts DPI through DP4 become roughly equal
to each other.
[0039] Then, each of the processes performed on and after the step S20 shown in Fig. 2 will
be explained. In the step S20, the detection image TP represented by the detection
image data output from the detection image generation section 232 of the detection
image adjustment section 230 to the image display processing section 340 is su
perimposed on an image represented by image data output from the input processing
section 320 of the image processing operation circuit 30 to the image display
processing section 340, and then displayed by projection on the screen SC via the light
valve drive section 380 and the image projection optical system 40 (Fig. 1). Then, in
the step S30, the imaging control section 234 (Fig. 1) of the detection image ad
justment section 230 controls the imaging section 50 to take the projection image
including the detection image and displayed by projection on the screen SC, and thus
the image data (also referred to as "taken image data") of the taken image is acquired,
and is then stored in the taken image information storage section 266.
[0040] In the step S40, the image analysis section 236 (Fig. 1) of the detection image ad
justment section 230 obtains the luminance of each of the detection image parts
(hereinafter also referred to as "taken detection image parts") in the taken image rep
resented by the taken image data, the detection image parts corresponding to the dot
patterns (the detection image parts) DPI through DP4 constituting the detection image
TP. Specifically, since approximate positions where the detection image parts re
spectively corresponding to the dot patterns DPI through DP4 must exist can be
identified in, for example, the taken image, the maximum luminance value of the
pixels within the approximate range including each of the detection image parts is
detected, and then the maximum luminance value thus detected is determined as the
luminance of the corresponding one of the detection image parts.
[0041] In the step S50, the image analysis section 236 judges whether or not all of the
luminance values (the maximum luminance values) of the detection image parts in the
taken image are within the allowable range to thereby determine the necessity of the
correction of the detection image. Specifically, the judgment can be achieved by de
termining whether or not all of the maximum luminance values of the detection image
parts in the taken image are within the range of 230 through 250 in an 8-bit gray-scale
(90 through 98 % in the expression with the proportion to the maximum gray-scale
value 255). It should be noted that the range is nothing more than an exemplification,
but is not necessarily limited thereto, and is sufficiently set to a range corresponding to
the required adjustment accuracy.
[0042] If all of the luminance values (the maximum luminance values) of the respective
detection image parts in the taken image thus obtained are within the allowable range,
and therefore, it is determined that the correction of the detection image is unnecessary
(YES in the step S50), the centroid coordinate detection section 238 (Fig. 1) of the
detection image adjustment section 230 obtains the centroid coordinates of the r e
spective detection image parts in the taken image corresponding to the dot patterns
DPI through DP4 as the detection image parts constituting the detection image TP in
the step S60a. Then, based on the centroid coordinates thus obtained here, various ad
justments such as the focus adjustment and the keystone correction are further
performed. It should be noted that the method of obtaining the centroid coordinates
will be explained later.
[0043] If any one of the luminance values (the maximum luminance values) of the r e
spective detection image parts in the taken image thus obtained is outside the allowable
range, and therefore it is determined that the correction of the detection image is
necessary (NO in the step S50), the detection image generation section 232 obtains the
correction information for correcting the detection image TP in the step S60b. Then,
the process returns to the step S10, the detection image is corrected based on the
correction information thus obtained, the image data of the detection image thus
corrected is output to the image display processing section 340, and then the process of
the steps S10 through S60b is repeated until it is determined in the step S50 that the
correction of the detection image is unnecessary. It should be noted that the process
(step S60b) of obtaining the correction information will be explained later.
[0044] As explained hereinabove, in the detection image adjustment process, the adjustment
of the detection image is performed so that the luminance values of the respective
detection image parts in the taken image become roughly equal to each other.
[0045] Method of Generating Detection Image
Fig. 5 is a flowchart showing a procedure of generating the dot patterns as the
detection image parts constituting the detection image in the step S10 shown in Fig. 2.
Firstly, determination of the dot pattern size sd (step S I 10), determination of the
number of gray levels stp of the luminance (step S120), and determination of the
standard deviation sigma (step SI 30) are performed. It should be noted that at the
beginning of the detection image adjustment, these parameters sd, stp, and sigma are
set to values determined in advance. In the explanation below, it is assumed that sd=34
[pixels], stp=10 [levels], and sigma=10 [pixels] are set, as an example of the initial
setting values.
[0046] Then, the values of the probability density A(0), A(sd/2) in s=0, and s=sd/2, r e
spectively, are calculated (step S140) from the normal distribution function A(s)
expressed by the formula (1) below, and then an allocation value Pa for each of the
levels of the number of gray levels stp is calculated (step S I 50) from the allocation
formula of the formula (2) below.
[Math.l]
[Math.2]
In the case of the initial setting of sd=34, stp=10, and sigma=10 described above,
A(0)=0.03989 and A(sd/2)=A(17)=0.00940 are obtained from the formula (1)
assuming that the average value ave is zero, and Pa=0.00305 is obtained from the
formula (2). These numerical values are each described to be rounded off to five
decimal places for the sake of convenience. It should be noted that although sd is
expressed within the range of -rn through +(rn-l), and the average value ave is -0.5
instead of 0 in the case in which the size sd of the dot pattern is an even pixel count as
explained with reference to Fig. 3B, since it is conceivable that there is hardly a com
putational problem if the range of -rn through -i-rn is assumed similarly to the case of
the odd pixel count, ave=0 is adopted as described above. For example, in the case of
sd=34, the difference in calculation value between the case of considering the actual
range of -17 through +16 and the case of considering the range of -17 through +17 is
about 0.00005 on the maximum value A(0) side, and about 0.00083 on the minimum
value A(17) side, which have no problem in treating them as roughly the same values.
Subsequently, the radius rn (n: 0 through stp-l=9) of each of the levels is calculated
(step S160). Specifically, the radius rn fulfilling the formula (3) below is calculated.
[Math.3]
[0049] In the case of the initial setting of sd=34, stp=10, and sigma=10 described above, the
radiuses rO through r9 of the respective levels are obtained as r0=4 [pixels], rl=6
[pixels], r2=7 [pixels], r3=9 [pixels], r4=10 [pixels], r5=ll [pixels], r6=12 [pixels],
r7=14 [pixels], r8=15 [pixels], and r9=17 [pixels].
[0050] Then, the regions of the respective levels are specified (step S170). Specifically,
defining the center coordinate of the dot pattern as the origin, the regions of the r e
spective levels are specified based on the formula (4) below. Specifically, the circle
with the radius rn expressed by the formula (4) forms a boundary between the regions,
and the inner area thereof is specified as each of the regions. Therefore, the area
between the circle expressed by (rn-l) 2=x2+y2 and the circle expressed by rn2=x2+y2
corresponds to the region of the n-th level.
[Math.4
(x: horizontal pixel position, y: vertical pixel position)
Lastly, the gray-scale value (the luminance value) Vn of the luminance of each of the
levels is set (step SI 80). Specifically, for example, it is possible to set the luminance
value Vn of each of the levels based on the formula (5) below.
[Math.5
[0052] In the case of the initial setting of sd=34, stp=10, and sigma=10 described above, the
luminance values V0 through V9 of the respective regions of the first level (the radius
rO) through the tenth level (the radius r9) can be obtained as, for example, VI: 88 %
white, V2: 78 % white, V3: 68 % white, V4: 58 % white, V5: 48 % white, V6: 38 %
white, V7: 28 % white, V8: 18 % white, and V9: 8 % white if V0: 98 % white (the
proportion to the maximum gray-scale value of 255 in 8 bits) is assumed.
[0053] Method of Obtaining Correction Information
Figs. 6A, 6B, and 7A through 7C are explanatory diagrams showing the method of
obtaining the correction information. Fig. 6A shows the detection image, Fig. 6B
shows a imaging result, Fig. 7A shows an example of correction concept, and Fig. 7B
shows a specific example of the correction content. It should be noted that the ex
planation will be presented assuming that the maximum luminance values Vdpl
through Vdp4 of the dot patterns (the detection image parts) DPI through DP4 are set
to 98 % (250 in 8 bits) to the maximum settable value of the luminance as shown in
Fig. 6A. It should be noted that the maximum luminance value Vdpl of the dot pattern
DPI, for example, is hereinafter also referred to simply as "luminance value Vdpl."
Further, in order to make the graphical description simpler, the dot patterns DPI
through DP4 of the detection image shown in Fig. 6A are each shown with the number
of levels of the luminance distribution of stp=5, which is lower than stp=10 in the
above exemplification. Further, the density of the hatching represents the magnitude
relation in luminance value between the levels, and specifically, it is expressed that the
higher the density of the hatching is, the lower the luminance value is, and the lower
the density of the hatching is, the higher the luminance value is.
[0054] It is assumed that the detection image shown in Fig. 6A is projected, and then the
detection image parts (the taken detection image parts) SDPl through SDP4 corre
sponding to the respective dot patterns DPI through DP4 are obtained as shown in Fig.
6B. As a result, the luminance value (the maximum luminance value) Vsdpl of the
taken detection image part SDPl located upper left is 75 % (191 in 8 bits), the
luminance value Vsdp2 of the taken detection image part SDP2 located upper right is
60 % (153 in 8 bits), the luminance value Vsdp3 of the taken detection image part
SDP3 located lower left is 100 % (255 in 8 bits), and the luminance value Vsdp4 of the
taken detection image part SDP4 located lower right is 80 % (204 in 8 bits). It should
be noted that as shown on the left of the taken detection image part SDP3 located
lower left, also in the case of the saturated extraction (brightly extracted), the
luminance value is 100 % (255 in 8 bits).
[0055] Here, in order to accurately obtain the centroid coordinate of each of the taken
detection image parts SDPl through SDP4 corresponding to the respective dot patterns
DPI through DP4 of the detection image, it is preferable that the luminance value (the
maximum luminance value) of each of the taken detection image parts SDPl through
SDP4 is equal to or higher than 90 % (230 in 8 bits). Further, if the luminance value is
100 % (255 in 8 bits), it could be the case of the saturated extraction as described
above. Therefore, it is preferable to set the upper limit of the luminance value to
around 98 % (250 in 8 bits). Therefore, it results that it is sufficient to correct the
luminance values Vdpl through Vdp4 of the respective dot patterns DPI through DP4
of the detection image so that the luminance values Vsdpl through Vsdp4 of the four
respective taken detection image parts SDPl through SDP4 fall within a predetermined
allowable range (90 through 98 % (230 through 250 in 8 bits)), and become roughly
equal to each other. It should be noted that the upper limit of the allowable range is not
necessarily required to be set to 98 % (250 in 8 bits), and if the saturated extraction
does not occur, the upper limit can be set to 100 % (255 in 8 bits).
[0056] Therefore, in the case of, for example, the imaging result shown in Fig. 6B, in order
to make the luminance values Vsdpl through Vsdp4 of the four taken detection image
parts SDPl through SDP4 fall within the allowable range and be roughly equal to each
other, the following will work from a simply conceptual viewpoint. That is, it is
possible to correct the luminance values Vdpl, Vdp2, and Vdp4 of the dot patterns
DPI, DP2, and DP4 located upper left, upper right, and lower right, respectively, into
the luminance values increased in accordance with how darkly the dot patterns are
extracted, and to correct the luminance value Vdp3 of the dot pattern DP3 located
lower left into the luminance value decreased in accordance with how brightly the dot
pattern is extracted to thereby make the luminance values Vsdpl through Vsdp4 of the
taken detection image parts SDPl through SDP4 corresponding respectively to the dot
patterns roughly equal to each other.
[0057] Here, in the case in which the luminance value Vadp2 of the corresponding taken
detection image part SDP2 is as low as 60 % as in the case of the dot pattern DP2
located upper right, it results that the correction into the luminance value roughly 1.6
times as high as the luminance value Vdp2 (98 ) set originally if following the
correction concept described above. However, since the luminance value approximated
to roughly 100 % is set, it results that the corresponding correction is not achievable by
simply raising the setting of the luminance value. Further, in the case in which the
luminance value Vsdp3 of the corresponding taken detection image part SDP3 is 100
% as in the case of the dot pattern DP3 located lower left, the saturated state is also
possible, and there is a possibility that it is not allowable to decrease the value simply
in accordance with how the dot pattern is extracted.
[0058] Incidentally, since the camera used in the imaging section 50 is generally provided
with an exposure adjustment function, the exposure is automatically controlled in ac
cordance with the luminance of the image. Therefore, it is known that even if the
luminance value of the taken image without the exposure adjustment is lower than 90
% (230 in 8 bits), the luminance value of the taken image is controlled to rise as a
whole due to the exposure adjustment. Therefore, it is conceivable that it is possible to
perform the correction so that the luminance value of each of the detection image parts
(the taken detection image parts) of the taken detection image falls within the
allowable range due to the effect of the exposure adjustment providing the luminance
values are corrected so as to become roughly equal to each other if the luminance
values of the respective taken detection image parts are lower than 90 %.
[0059] Therefore, in the present embodiment, the following correction is performed. That is,
as shown in Fig. 7B, it is possible that, taking the taken detection image part, which is
the most darkly extracted, as a reference, the luminance value (the maximum
luminance value) of the dot pattern corresponding thereto is set, and the luminance
values (the maximum luminance values) of the dot patterns corresponding to the taken
detection image parts, which are more brightly extracted than the reference, are set to
be darker in accordance with the relationships with the luminance value (the maximum
luminance value) of the taken detection image part, which is the most darkly extracted,
and thus, the luminance values (the maximum luminance values) of the respective
taken detection image parts are made to be roughly equal to each other. Specifically,
taking, for example, the dot pattern DP2 corresponding to the taken detection image
part SDP2, which is located upper right of Fig. 6B, and is the most darkly extracted, as
a reference, the luminance value Vdp2 thereof is corrected into 100 % so as to be the
brightest. The luminance value Vdp3 of the dot pattern DP3 corresponding to the taken
detection image part SDP3, which is located lower left, and is the most brightly
extracted, is corrected into 60 % so as to be darker than the reference in accordance
with the difference between the luminance value Vsdp3 of the taken detection image
part SDP3 corresponding thereto and the luminance value Vsdp2 of the taken detection
image part SDP2, which is taken as the reference. The luminance valued Vdpl, Vdp4
of the dot patterns DPI, DP4 corresponding to the taken detection image parts SDP1,
DSP4, which are located upper left and lower right, respectively, and are extracted
with intermediate darkness, are corrected into 80 % so as to be slightly darker than the
reference in accordance with the differences of the luminance values Vsdpl, Vsdp4 of
the taken detection image parts SDP1, SDP4 from the luminance value Vsdp2 of the
taken detection image part SDP2, which is taken as the reference. It should be noted
that a specific example of a method of obtaining the correction values will be
explained below.
[0060] Figs. 8A and 8B are explanatory diagrams showing a procedure of obtaining
correction values used for the correction of the detection image. Fig. 8A shows a
flowchart for obtaining the correction values, and Fig. 8B shows a specific example
executed with the procedure shown in Fig. 8A in the case in which the imaging result
shown in Fig. 6B is obtained. As shown in Fig. 8A, firstly, the luminance values Vsdpl
through Vsdp4 (step S40 in Fig. 2) of the detection image parts (the taken detection
image parts) SDP1 through SDP4 corresponding to the respective dot patterns DPI
through DP4 having already been obtained are obtained (step S210). Specifically, for
example, Vsdpl=75 %, Vsdp2=60 %, Vsdp3=100 %, and Vsdp4=80 % are obtained as
shown in Fig. 8B.
[0061] Subsequently, using the formula (6) below, the minimum value Vmin of the four
luminance values, and the proportions Km (m: 1 through 4) of the respective
luminance values are calculated (step S220). Specifically, as shown in Fig. 8B, Kl=80
%, K2=100 %, K3=60 %, and K4=75 % are obtained.
[Math.6]
Km = (Vmin/Vsdpm) -100 ( )
[0062] Then, by rounding the proportions Km thus obtained to the nearest 10, the correction
values Cm (m: 1 through 4) are obtained (step S230). Specifically, as shown in Fig.
8B, Cl=80 %, C2=100 %, C3=60 %, and C4=80 % are obtained. It should be noted
that the process of rounding the proportions to the nearest ten is for reducing the com
putation load, and can therefore be changed in accordance with the computation load.
For example, a process of rounding the proportions off to the nearest whole number
can also be adopted.
[0063] It should be noted that as a result, the detection image generation section 232
performs the correction of the dot patterns based on the correction values (the
correction information) obtained in such a manner as described above. In the imaging
result corresponding to the detection image thus corrected, the luminance values Vsdpl
through Vsdp4 of the taken detection image parts SDPl through SDP4 each fall within
the allowable range (230 through 250 in 8 bits (90 through 98 )) as shown in Fig. 7C.
[0064] As the correction method of the dot patterns, a variety of methods can be adopted.
For example, in the case of generating the dot patterns as explained with reference to
Fig. 5, it is possible to generate the dot patterns by changing the parameters thereof
such as the dot pattern size sd, the number of gray levels stp, the standard deviation
sigma, and the luminance value V0 of the center region. If the dot size sd is increased,
the range of application of the normal distribution is increased to thereby increase the
distribution quantities of the respective levels. Therefore, the maximum luminance
value of the taken detection image part tends to be raised. In contrast, if the dot size sd
is decreased, the range of application of the normal distribution is decreased to thereby
decrease the distribution quantities of the respective levels. Therefore, the maximum
luminance value of the taken detection image part tends to be lowered. If the number
of gray levels stp is increased, the width of each of the levels is decreased, and
therefore, the maximum luminance value of the taken detection image part is
decreased. If the number of gray levels stp is decreased, the width of each of the levels
is increased, and therefore, the maximum luminance value of the taken detection image
part tends to be increased. If the standard deviation sigma is increased, the normal dis
tribution becomes gentler to increase the width of the center region, and thus, the
maximum luminance value of the taken detection image part tends to be increased. In
contrast, if the standard deviation sigma is decreased, the normal distribution becomes
steeper to decrease the width of the center region, and thus, the maximum luminance
value of the taken detection image part tends to be decreased. If the luminance value
V0 of the center region is increased, the maximum luminance value of the taken
detection image part is increased, and if the luminance value V0 of the center region is
decreased, the maximum luminance value of the taken detection image part is
decreased. Therefore, by appropriately setting the values of these parameters in ac
cordance with the correction values described above to thereby perform the correction
of the dot patterns, it becomes possible to correct the dot patterns so that the imaging
result with the desired maximum luminance values (the luminance values) can be
obtained.
[0065] It should be noted that although the taken detection image part with the lowest
luminance is taken as the reference in the explanation of the example of obtaining the
correction values described above, it is also possible to arrange that the taken detection
image part with intermediate luminance or the taken detection image part with the
highest luminance is taken as the reference, and the luminance distributions of the dot
patterns as the corresponding detection image parts are changed so that the luminance
values of the respective taken detection image parts become roughly equal to each
other.
[0066] Method of Obtaining Centroid Coordinate
Figs. 9A and 9B are explanatory diagrams showing a procedure of obtaining the
centroid coordinate in the step S60a shown in Fig. 2. Fig. 9A shows a flowchart of the
procedure of obtaining the centroid coordinate, and Fig. 9B shows a target region of
the centroid coordinate calculation.
[0067] As shown in Fig. 9A, the taken image data is read (step S3 10), and then extraction of
the centroid coordinate calculation target region from the taken image data thus read is
performed (step S320). Specifically, the process is performed as described below, for
example. Since the dot pattern (the detection image part) constituting the detection
image is specified with a coordinate, an approximate area Aex such as an area with a
half or quarter width or height from the four corners can be identified even in the taken
image represented by the taken image data as shown in Fig. 9B. Therefore, the
maximum luminance value in each of the areas Aex is detected. Further, it is possible
to extract the minimum area including the corresponding taken detection image part as
the centroid coordinate calculation target region Ag based on the coordinate of the
maximum value and the dot pattern size. Then, the calculation of the centroid co
ordinate is performed (step S330) in each of the centroid coordinate calculation target
regions Ag thus extracted.
[0068] Fig. 10 is a flowchart showing a procedure of centroid coordinate calculation in one
centroid coordinate calculation target region. Firstly, the taken image data in the
centroid coordinate calculation target region Ag is investigated to obtain (step S410)
the maximum value Vmax and the minimum value Vmin of the luminance in the
region. Further, the threshold value th is obtained (step S420) using the formula (7)
below.
[Math.7]
th = (V max- Vmin) · 0.25 + V min · ·(7)
[0069] It should be noted that the formula (7) shows that the value, which is 25 % of the
difference (Vmax-Vmin) greater than the minimum value Vmin of the luminance in
the centroid coordinate calculation target region Ag, is set to the threshold value. It
should be noted that the value of percentage of the difference (Vmax-Vmin) is not
limited to 25 , but can arbitrarily be set in accordance with what is the value set to
the minimum luminance of the pixel to be the centroid coordinate calculation target in
the centroid coordinate calculation target region Ag.
[0070] Then, the comparison between the luminance value V(x, y) of each of the pixels of
the centroid coordinate calculation target region Ag and the threshold value th is
performed, and if V(x, y)-th>0 is fulfilled, it is determined that the pixel is included in
the target of the centroid coordinate calculation, and the accumulating operations
expressed in the formulas (8) through (10) below are performed (step S430). The
formula (8) means that the luminance values of the pixels determined to be the target
of the centroid coordinate calculation are accumulated. The formula (9) means that the
product of the x-coordinate value and the luminance value of the pixel determined to
be the target of the centroid coordinate calculation is accumulated. The formula (10)
means that the product of the y-coordinate value and the luminance value of the pixel
determined to be the target of the centroid coordinate calculation is accumulated. It
should be noted that the process is repeated (step S440) until the process is performed
on all of the pixels in the centroid coordinate calculation target region Ag.
[Math. 8]
Sum - Sum + V(x, y) · · · (8)
[Math.9]
SumX = SumX + [V(x, y) x] (9)
[Math. 10]
SumY = SumY + [V(x, y) y] (10)
[0071] Then, if the process of the step S430 has been performed on all of the pixels in the
centroid coordinate calculation target region Ag (YES in the step S440), whether or not
the value of the parameter Sum is 0 is determined (step S450). If the value of the
parameter Sum is 0 (NO in the step S450), it is determined that an error occurs in the
calculation of the centroid coordinate (xg, yg), and an error value determined in
advance is set as the centroid coordinate (xg, yg). It should be noted that in this case it
is also possible to arrange that the flow of obtaining the centroid coordinate shown in
Fig. 9A is resumed to thereby achieve reduction of the rate of occurrence of the error.
In contrast, if the value of the parameter Sum is not equal to 0 (NO in the step S450),
the centroid coordinate (xg, yg) is obtained with the formulas (11), (12) below.
[Math. 11]
[Math. 12]
[0072] It should be noted that the formula (11) means that the x-coordinate value of the
centroid is obtained by dividing the integrated value of the product of the x-coordinate
value and the luminance value of each of the pixels determined to be the target of the
centroid coordinate calculation by the integrated value of the luminance value of each
of the pixels determined to be the target of the centroid coordinate calculation.
Similarly, the formula (12) means that the y-coordinate value of the centroid is
obtained by dividing the integrated value of the product of the y-coordinate value and
the luminance value of each of the pixels determined to be the target of the centroid co
ordinate calculation by the integrated value of the luminance value of each of the
pixels determined to be the target of the centroid coordinate calculation.
[0073] Detection Image Correction Effect
The effect of the detection image correction explained with reference to Figs. 2, 3A,
3B, 4A, 4B, 5, 6A, 6B, 7A through 7C, 8A, and 8B is confirmed. Specifically, in the
case in which the projector PJ (Fig. 1) according to the present embodiment is installed
in front of the screen SC with the projection distance therebetween set to 70 cm, the
focus adjustment is performed, and then the projection distance is changed to 300 cm
to create the defocused condition, the centroid coordinate calculated in the case of not
correcting the detection image, and the centroid coordinate calculated in the case of
correcting the detection image are evaluated. It should be noted that the dot patterns of
the detection image before the correction are generated assuming that the dot pattern
size sd is 34 [pixels], the number of set gray levels stp is 10 [levels], the standard
deviation sigma is 10 [pixels], and the luminance value V0 of the center region is 250
(98 ) in 8 bits, similarly to the case exemplified in the explanation of Fig. 5.
[0074] Fig. 11 is a table showing the evaluation results with the detection image correction
and the evaluation results without the detection image correction in a comparative
manner. Fig. 12 is a plurality of graphs, each showing a variation in the centroid co
ordinate over a plurality of times of measurement for the results with the detection
image correction and without the detection image correction, arranged in a com
parative manner. As shown in Fig. 11, there is obtained a result that the luminance
values (the maximum luminance values) of the taken detection image parts corre
sponding to the respective dot patterns DPI through DP4 of the detection image before
the correction are 65 , 77 , 87 , and 100 , respectively. It should be noted that
the result shows a representative example obtained over a plurality of times (e.g., 100
times) of measurement. Then, the dot patterns are corrected by setting the value of the
standard deviation sigma to sigma=9 in the dot pattern DPI located upper left and the
dot pattern DP3 located lower left, and to sigma=8 in the dot pattern DP2 located upper
right and the dot pattern DP4 located lower right based on the result described above.
As a result, as shown in Fig. 11, there is obtained a result that the luminance values of
the taken detection image parts corresponding to the respective dot patterns DPI
through DP4 of the detection image after the correction are 100 , 92 , 100 , and
90 , respectively. It should be noted that the result also shows a representative
example.
[0075] Further, as shown in Fig. 12, the variation in the centroid coordinate obtained by
performing the imaging and then measuring the image a plurality of times (here, 100
times) can be reduced compared to the case before the correction. In particular, as is
obvious from the result of the dot pattern DPI located upper left with the remarkably
low proportion of the luminance value before the correction, it is understood that the
proportion is dramatically improved, and the difference in variation between the four
dot patterns is reduced. Similarly, as is obvious from the result of the dot pattern DPI
located upper left with the remarkably low proportion of the luminance value, it is u n
derstood that the sum of the shift amount of the centroid coordinate and the standard
deviation sigma of the centroid coordinate measured in Fig. 11 are also dramatically
improved, and the difference in variation between the four dot patterns is reduced.
[0076] It can be confirmed that by performing the correction so that the luminance values of
the taken detection image parts corresponding to the respective dot patterns fall within
the allowable range, and become roughly equal to each other as described above, the
extraction accuracy of the centroid coordinate in each of the points is improved.
[0077] As explained hereinabove, in the projector according to the present embodiment, by
correcting the detection image so that the luminance values of the detection image
parts (the taken detection image parts) in the taken image corresponding to the dot
patterns as the detection image parts of the detection image fall within the allowable
range and become roughly equal to each other, the detection accuracy of the detection
image can be improved. Further, even in the state in which the focus adjustment is not
performed to thereby cause defocus, or the state in which the keystone correction is not
performed to thereby cause a distortion in the projection image, the detection accuracy
of the detection image can similarly be improved. As a result, it becomes possible to
perform a variety of adjustments such as the focus adjustment and the keystone
correction with good accuracy using the detection image extracted with accuracy.
[0078] C. Modified Examples
It should be noted that the invention is not limited to the embodiment described
above, but can be put into practice in various forms within the scope of the invention.
[0079] 1. First Modified Example
In the embodiment described above, the explanation is presented citing the case as an
example in which the difference in probability density between the levels is set to have
a constant interval using the allocation value determined using the normal distribution
function as shown in Fig. 5, and at the same time the luminance values of the r e
spective levels are set so that the ratios between the luminance values decrease at
regular intervals in the generation of the dot patterns as the detection image parts.
However, the invention is not limited thereto, but it is also possible to use a linear
function with a broken line shape, a quadratic function, or the like instead of the
normal distribution function. Further, it is possible to set the luminance values of the
respective levels so that the gray-scale values decrease at regular intervals instead of
the ratios between the luminance values with the regular intervals, and the regular
intervals in the gray-scale values or the ratios are not necessarily required.
[0080] Further, although in the embodiment described above, the case of changing the
standard deviation sigma is explained as an example of the correction of the dot
patterns, it is also possible to arrange that the correction is performed by changing a
variety of parameters such as the dot pattern size, the number of gray levels, or the
luminance value of the center region as explained in the description of the em
bodiment.
[0081] As described above, the generation and the correction of the detection image parts as
the dot patterns can be performed using any method providing the detection image
parts each including a plurality of regions having the respective luminance values
different form each other can be generated and corrected.
[0082] 2. Second Modified Example
As shown in Fig. 3A, in the embodiment described above, the explanation is
presented citing the detection image having the four dot patterns disposed at the four
corners of the image as the detection image parts as an example. However, the
invention is not limited thereto, but a variety of detection images can be used.
Hereinafter, some other detection images will be cited.
[0083] Figs. 13A and 13B are explanatory diagrams each showing an example of another
detection image using the dot patterns. Fig. 13A shows an example having nine dot
patterns disposed in a grid. Fig. 13B shows an example in which four blocks each
including four dot patterns are disposed in the four corners of the image, wherein each
of the blocks is composed of four dot patterns arranged at vertexes of a square. Such
detection images having the dot patterns as the detection image parts variously
changed in the number, the arrangement, and so on as described above can also be
used.
[0084] Fig. 14 is an explanatory diagram showing an example of the detection image
including other detection image parts different from the dot pattern. Fig. 14 shows an
example of a line image shaped like a grid. The line image is sectioned into a plurality
of regions having respective luminance values different from each other arranged so
that the center region of the line has a higher luminance value and the outer region of
the line has a lower luminance value. In the case of the detection image, it is possible
to use the parts indicated by the circular frames as the detection image parts. The
correction can be performed by varying the width of the line, the number of gray
levels, the width of each level, the setting values of the luminance of the center region,
and so on. As described above, the detection image part is not limited to the dot
pattern, but a variety of detection images can be used providing the detection image
has a plurality of detection image parts each having a plurality of regions with r e
spective luminance values different from each other.
[0085] 3. Third Modified Example
Fig. 15 is an explanatory diagram showing a configuration of displaying images on a
projection surface in a superimposed manner using a plurality of projectors. It should
be noted that Fig. 15 shows the case in which the projection images from two
projectors PJ1, PJ2 are superimposed on the screen SC as the projection surface, and
thus a single image is displayed as an example. The configuration of each of the two
projectors PJ1, PJ2 is substantially the same as in the embodiment.
[0086] In the case of displaying one image by superimposing the projection images from a
plurality of projectors, in order to display the images while accurately superimposing
them on each other, it is important to extract the detection images with good accuracy,
and to accurately obtain the centroid coordinate of each of the taken detection image
parts corresponding to the detection image parts of the detection image. In the example
shown in Fig. 15, since the two projectors PJ1, PJ2 each having substantially the same
configuration as that of the projector PJ explained in the description of the embodiment
are used, it becomes possible to accurately extract the detection image in each of the
projectors to thereby accurately obtain the centroid coordinate of the taken detection
image parts corresponding to the detection image parts of the detection image as
explained in the description of the embodiment. Thus, it is possible to improve the
accuracy of the superimposition of the images.
[0087] It should be noted that it is also possible to make the detection image different
between the first projector PJ1 and the second projector PJ2. According to this con
figuration, the detection images projected on the screen SC can easily be separated,
and thus the extraction of the detection image in each of the projectors becomes easy.
In order to make the detection image different, there can be cited a variety of methods
such as change in the layout or the number of the dot patterns, or change in the pattern
itself of the detection image between the dot pattern shown in Figs. 3A, 13A, and 13B
and the pattern shown in Fig. 14. Further, the invention is not limited to two projectors,
but is advantageous in the case of displaying one image by superimposing the
projection images from a plurality of projectors.
[0088] 4. Fourth Modified Example
The method for obtaining the correction values explained with reference to Figs. 6A,
6B, 7A through 7C, 8A, and 8B, and the formula used therefor in the embodiment
described above are illustrative only, but the invention is not limited thereto. Any
method and formula can be used providing the luminance values of the respective
detection image parts (the taken detection image parts) of the taken image fall within
the allowable range and become roughly equal to each other using the method and the
formula.
[0089] 5. Fifth Modified Example
In the embodiment described above, the calculation of the centroid coordinate is not
necessarily limited to those using the formulas (11) and (12), but a variety of cal
culation methods of the centroid coordinate can be used. For example, it is also
possible to adopt the average value of the coordinates of the pixels each having a
greater luminance value than the threshold value th. Further, it is also possible to adopt
the average value of the coordinates of the pixels located in the centroid coordinate cal
culation target region Ag.
[0090] 6. Sixth Modified Example
In the embodiment described above, it is assumed that the detection image ad
justment is started in response to the instruction issued by the user operating the input
operation section 10. However, the invention is not limited thereto, but the detection
image adjustment can be started at various timings. For example, it is possible to start
the detection image adjustment automatically when starting up the projector. Further, it
is also possible to arrange that the detection image adjustment is started automatically
in response to the motion detection section 60 detecting the transition of the state of the
projector from a resting state to a moving state.
[0091] 7. Seventh Modified Example
Although the explanation is presented assuming that the process is terminated after
obtaining the centroid coordinate of each of the detection image parts in the detection
image adjustment process (Fig. 2) of the embodiment described above, it is also
possible to arrange that, for example, the process keeps the standby state until the ter
mination instruction from the user is issued, and performs the adjustment of the
detection image once again if it is determined in the standby state that the readjustment
of the detection image is necessary due to the fact that the motion detection section
detects the motion of the projector is detected and the change in the installation state is
detected, or the fact that the setting condition of the projector is changed. It should be
noted that it is also possible to arrange that the process is terminated if the read
justment of the detection image fails to become necessary for a predetermined period
of time instead of the termination instruction from the user.
[0092] 8. Eighth Modified Example
In the embodiment described above, the explanation is presented citing the case of
providing the projector with the imaging section an example. However, it is also
possible to arrange that the imaging section is provided separately from the projector.
Also in this case, the detection image can be corrected in accordance with the image
taken by the imaging section provided separately from the projector. Thus, it becomes
possible to extract the detection image with accuracy to thereby accurately obtain the
centroid coordinate of each of the taken detection image parts corresponding to the
detection image parts of the detection image. Thus, it is possible to improve the
accuracy of the superimposition of the images.
[0093] 9. Ninth Modified Example
In the embodiment described above, the projector PJ converts the light from the illu
mination optical system 420 into the image light using the liquid crystal light valve
440, which uses the transmissive liquid crystal panel, as the light modulation device.
However, it is also possible to use a digital micro-mirror device (DMD), a reflective
liquid crystal panel, or the like as the light modulation device.
Reference Signs List
[0094] 10 input operation section
20 control circuit
30 image processing operation circuit
40 image projection optical system
50 imaging section
220 Control section
230 detection image adjustment section
232 detection image generation section
234 imaging control section
236 image analysis section
238 centroid coordinate detection section
260 information storage section
262 setting information storage section
264 detection image information storage section
266 taken image information storage section
320 input processing section
340 image display processing section
360 image memory
380 light valve drive section
420 illumination optical system
422 light source lamp
424 lamp drive section
440 liquid crystal light valve
460 projection optical system
462 projection lens
464 lens drive section
466 state detection section
PJ projector
PJ1, PJ2 projector
TP, TPa detection image
PP image
DP dot pattern
DPI through DP4 dot pattern (detection image part)
SDPl through SDP4 taken detection image part
SC screen (projection surface)
PCT/JP2013/001880
Claims
An image processing device used for a projector adapted to display an
image by projecting the image on a projection surface comprising:
a detection image generation section adapted to generate a detection
image, which is an image adapted to detect a state of a projection image
displayed on the projection surface, and includes a plurality of
detection image parts,
wherein each of the detection image parts includes a plurality of
regions having respective luminance values different from each other,
and
the detection image generation section changes a luminance dis
tribution of each of the detection image parts of the detection image to
be generated so that maximum luminance values of the detection image
parts included in a taken detection image obtained by taking the
detection image projected on the projection surface fall within an
allowable range.
The image processing device according to Claim 1, wherein
the detection image generation section takes the luminance value of
either one of the taken detection image parts out of the detection image
parts included in the taken detection image as a reference luminance
value, and changes the luminance distribution of each of the detection
image parts of the detection image to be generated so that the
maximum luminance values of the detection image parts included in
the taken detection image are roughly equal to each other based on a r e
lationship between the reference luminance value and luminance values
of other detection image parts.
The image processing device according to Claim 1 or 2, wherein
the detection image generation section changes the luminance dis
tribution of each of the detection image parts of the detection image to
be generated by changing at least one of
an overall size of the detection image part included in the detection
image,
a width of each of the regions included in the detection image part, and
the maximum luminance value of the detection image part.
The image processing device according to any one of Claims 1 through
3, wherein
sectioning of the regions is obtained based on a predetermined function.
PCT/JP2013/001880
The image processing device according to Claim 4, wherein
the function is a Gaussian distribution function.
The image processing device according to any one of Claims 1 through
5, further comprising:
a projection image adjustment section adapted to adjust image quality
of the projection image,
wherein the projection image adjustment section performs at least one
of a plurality of image quality adjustments including a focus adjustment
and a keystone distortion correction of the projection image based on
the process using the detection image.
A projector adapted to display an image by projecting the image on a
projection surface, comprising:
the image processing device according to any one of Claims 1 through
6;
an imaging section adapted to take the detection image projected on the
projection surface; and
a projection section adapted to project the image based on image data
output from the image processing device.
A method of controlling a projector adapted to display an image by
projecting the image on a projection surface, the method comprising:
(a) generating a detection image, which is an image adapted to detect a
state of a projection image displayed on the projection surface, and
includes a plurality of detection image parts; and
(b) projecting the detection image on the projection surface,
wherein each of the detection image parts includes a plurality of
regions having respective luminance values different from each other,
and
in step (a), a luminance distribution of each of the detection image parts
of the detection image to be generated is changed so that maximum
luminance values of the detection image parts included in a taken
detection image obtained by taking the detection image projected on
the projection surface fall within an allowable range.