FORM 2
THE PATENTS ACT, 1970
(39 of 1970)
&
THE PATENT RULES, 2003
COMPLETE SPECIFICATION
(See Section 10 and Rule 13)
Title of invention:
A SYSTEM AND METHOD FOR VIDEO SIGNALS ANALYSIS
APPLICANT:
TATA Consultancy Services Limited A company Incorporated in India under The Companies Act, 1956
Having address:
Nirmal Building, 9th Floor,
Nariman Point, Mumbai 400021
Maharashtra, India
The following specification particularly describes the invention and the manner in which it is to be performed.

FIELD OF THE INVENTION:
The present invention relates to analysis of video signals and particularly to the analysis of broadcasted video signals comprising plurality of data signals bundled in a broadcast stream.
PRIOR-ART REFERENCES:
1. Hua, X.S., Lu, L. and Zha, H.J., Robust Learning based TV commercial Detection, Proceedings of ICME 2005.
2. Sadlier, D.A.,et al. Automatic TV Advertisement Detection from MPEG bitstream, Pattern Recognition, Volume 35, Issue 12, 2002.
3. Gauch, J.M. and Shivadas, A. Identification of new commercials using repeated video sequences detection. Proceedings of ICIP 2005.
4. Liu Zhen, Sch. of Computer Sci. & Inf. Technology., Zhejiang Wanli Univ., Ningbo, China - "Commercial Detection in Program Videos", Vol. 3, 25-27 Dec. 2009.
5. En-Hao Yeh et al. Dept. of Comput. Sci. & Inf. Eng., Nat. Taiwan Univ., Taipei, Taiwan -"TV commercial detection in news program videos", Vol. 5,23-26 May 2005.
BACKGROUND OF THE INVENTION:
In broadcasted video signals e.g. news or sports videos, commercials are often inter-mixed with respective regular and premium contents of news or sporting events. For efficient analysis which includes retrieval and browsing of the premium video signals, post broadcasting thereof, detection and removal of unwanted or irrelevant contents e.g. commercials is generally desirable.
Hitherto, it is a very common practice in the media industry to use black/blank frames to detect commercials. However, the black/blank frames detection based approach is not reliable and gives erroneous results particularly for videos of TV channels that do not use black frames to flag commercial breaks. Also, black frames used in other parts of the broadcast will cause false alarms. Furthermore, progress in digital technology obviates the need to insert black frames before commercials during production. These approaches generally depends on the 'high

activity' rate which may not always distinguish commercials from regular broadcasts.
The existing technologies generally identify commercials by a combination of several features, such as change in pace in the video, change in audio characteristics, appearance of logos. The dynamic nature of video broadcasting and change that has occurred in production technology over recent years makes some of these methods unusable. Moreover, use of multimodal features makes existing methods compute-intensive.
In one of the cases wherein the news videos have evolved a lot and include plurality of information contents for viewers' consumption, the existing methods for commercial detection are somewhat not effective. The various publications and patents suggest use of video features based on the predefined video specifications. This method is vulnerable to similar video specification. A blank/black frames presence, absence of channel logo etc. cannot be used as indicators for advertisement detection these days as they are not form the part of video specification during advertisements.
Particularly, in news videos various news parts are presented using animations and editing. In the present scenario especially in India during news there are a lot of hard cuts in a story itself and audio type also varies during the news. This is done for various reasons: to emphasize the story by reciting facts about it in flashy way, editor's comments and news headlines are shown this way to attract viewer's attention, news headlines are shown many times in a news program, etc.
Hence, a real-time system and method is required for commercial detection that can be universally applicable to news, sport and other similar broadcast channels in any language.
Other features and advantages of the present invention will be explained in the following description of the invention having reference to the appended drawings.

OBJECTS OF THE INVENTION:
A primary object of the invention is to provide a system and method for detecting a commercials/ advertisements in a broadcast videos signals using change in Ticker Text positions.
Another object of the invention is to provide a system and method for checking the position and appearance of this global ticker text band to identify a possible advertisement segment.
Another object of the invention is to provide a system and method for improving the efficiency of a broadcasted video signals analysis system by way of reducing the computation and storage requirements for the commercial part of the news video.
SUMMARY OF THE INVENTION:
The present application provides a system and method for detecting the commercials/ advertisements in plurality of video signals that are generally broadcasted and comprises of primary regular signals and a secondary commercial or miscellaneous signals. The present application enables a detection of secondary signals facilitating post broadcast processing and analysis of the video signals. The present invention continuously monitors the change in Ticker Text positions in the broadcasted video signals. Particularly, it helps to improve the efficiency of a news analysis system by way of reducing the computation and storage requirements for the commercial part of the news video.
Further, the present application is configured to determine the region of interest in the broadcasted video signals and subsequently initiate the recording the instance of change in the position of the global ticker text band or its removal during the advertisements for creating more space for the advertisement content. The invention adapted to extract the broadcasted video signals in terms of differential frames thereof and to detect at least one frame for change in the position and appearance of this global ticker text band to identify a possible advertisement segment.

The commercial/advertisement detection improves the efficiency of a news analysis system as it reduces the computation and storage requirements of the commercial part of the news video, which is quite significant. After removing commercials, better analysis of news can be done, as commercial would not be able to produce any noise for the news analysis system.
BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
The foregoing summary, as well as the following detailed description of preferred
embodiments, is better understood when read in conjunction with the appended drawings. For
the purpose of illustrating the invention, there is shown in the drawings example constructions
of the invention; however, the invention is not limited to the specific system and method
disclosed in the drawings:
Figure 1 is a block diagram of the present invention according to one embodiment of the
invention.
Figure 2 is a block diagram of marking of Advertisement regions with "Ad Start" and "Ad End" tags in time stamped news video frames.
Figure 3 is a pictorial representation of actually captured television broadcast illustrating different segments of news and commercial advertisements in the broadcasted video.
Figure 4 is a pictorial representation of actually captured television broadcast depicting comparative variation in different regions in a broadcasted news and advertisement video.
DETAILED DESCRIPTION OF THE INVENTION
Some embodiments of this invention, illustrating its features, will now be discussed in detail. The words "comprising," "having," "containing," and "including," and other forms thereof, are intended to be equivalent in meaning and be open ended in that an item or items following any one of these words is not meant to be an exhaustive listing of such item or items, or meant to be limited to only the listed item or items.

It must also be noted that as used herein and in the appended claims, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. Although any methods, and systems similar or equivalent to those described herein can be used in the practice or testing of embodiments of the present invention, the preferred methods, and systems are now described.
The disclosed embodiments are merely exemplary of the invention, which may be embodied in various forms.
The news videos have this property that during the advertisements, Local ticker text bands is not present and static Global ticker text band shifts down or goes away completely. The absence of local ticker text is not limited to advertisements only. It can be absent in other parts of news videos too like during anchor shots or during the start and end of a news story. But the global ticker text is uniformly present during the news and shifts down or completely washed away during advertisements.
The present invention is based on the observation that global band is either shifts down or not present during the advertisements. So, a user needs to locate the positions in the news video, where global ticker text either shifts down or goes away. For this we first select a region in video frames which is upper half of the global ticker text band, as this can cover both the possibilities: shifting of the band and complete absence. Variation of Global ticker text is measured by presence of text in that region. The assumption here is that location of global ticker text band is already known.
For text detection we are using horizontal edge densities in an image. These are extracted using Sobel operator. Sobel operator is a differentiation operator, which computes an approximation of the gradient of the image intensity. The region having text will have more number of horizontal edges. Based on the presence of the ticker text, decision regarding advertisements is being taken. The various steps of the algorithm for Commercial detection are explained point wise below,
According to one embodiment of the invention, the primary signals correspond to the primary or regular video signals broadcasted for viewers' consumption and the secondary signals

corresponds to commercial contents broadcasted along with the primary signals.
According to one embodiment of the invention, the instance of starting of the secondary signals correspond to occurrence of one or more non text frames detected is an indication of "Ad Start" and the instance of ending of the secondary signals correspond to occurrence of one or more text frames detected is an indication of "Ad End".
Referring to Fig. 1, the commercial detection system comprises frame extraction module 10, Region of Interest module 20, Binarization module 30, an edge detection module 40, clustering module 50, commercial block detection module 60 comprises an "Ad Start" region and an "Ad End" region and post processing module 70.
Video frames are first extracted (10) from the given news video. These frames are then used for further processing.
Let I„ be the nth video frame where n=I, 2, , N, where N is the total number of frames in the
news video. The Region of interest (ROI) Rn in nth frame is chosen to be the upper half of the global ticker text band in a video frame. Let (xl, yl) and (x2, y2) are the coordinates of the upper left corner and lower right corner coordinates of the global ticker text band respectively. Then the coordinates for the ROI will be: (xl, yl) and (x;2, (yl + y2) / 2). The reason for choosing the ROI (20) is because it will contain either no text or very little text during advertisement. In is then cropped, to get the region of interest Rn.
The cropped region Rn is then binarized (30) and hence produced a black & white image "Bn" which is also known as monochrome image. The monochrome image Bn is then supplied to the edge detection module (40) and hence created an edge image "En". The edge detection is done using the Sobel operator. The number of horizontal edges of the En is counted which is then used to take decision on the presence of text or non text in the given region.
To decide, whether the ROI has text or not, number of horizontal edges in all the Ens is supplied to k-means clustering module (50) which is basecl upon an existing algorithm. The output of the k-means will be two clusters. The cluster that contains the regions having lesser number of horizontal edges is the one which doesn't have; text in the ROIs. And the other

Cluster that contains the regions having more number of horizontal edges is the another one which have texts.
Further, for commercial detection (60) a sliding window of M frames is used. This is set on the basis of the frame rate (fps) at which frames from the video are extracted and the minimum observed duration (Dmin) for a single block of commercials. Starting from the first frame, the sliding window is applied to each frame. The frame where the sliding window counts more than Mth, frames are having Rn's classified as non text regions, is marked as "Ad Start". Next time whenever the count of non text regions in the sliding window becomes less than Mth, that frame is marked as "Ad End". This way an array is generated which contains start and end times of the advertisement regions.
In final step, the post processing (70) is done on the basis of simple heuristic that two different commercial blocks cannot appear within the gap of Gth frames. If they have appeared in our case they are merged because there is a possibility in some commercial that the region of interest (ROI) is having information about a product in text form.
According to another embodiment of the invention, at least one set of secondary signals wherein the edge detection means is configured to concatenate plurality of disparate commercial video blocks detected with a presence of text in the region of interest encountered by the sliding window frame within the gap of predetermined minimum number of frames.
Fig. 2 depicts the marking of Advertisement regions with "Ad Start" and "Ad End" tags in time stamped news video frames in accordance with one general embodiment of the present application.
There are various frames that are extracted from news video is shown in Fig.3. A sliding window of M frames is used for commercial detection. This is set on the basis of the frame rate (fps) at which frames from the video are extracted and the minimum observed duration (Dmin) for a single block of commercials. Starting from the first frame, the sliding window is applied to each frame. The frame where the sliding window counts more than Mth frames are having Rn's classified as non text regions, is marked as "Ad Start" which is shown in dark black frame. Next time whenever the count of non text regions in the sliding window becomes less

than Mth, that frame is marked as "Ad End" which is also shown in dark black frame. This way an array is generated which contains start and end times of the Advertisement regions.
According to one embodiment of the invention, the instance of starting of the secondary signals correspond to occurrence of one or more non text frames detected is an indication of "Ad Start" and the instance of ending of the secondary signals correspond to occurrence of one or more text frames detected is an indication of "Ad End".
According to another embodiment of the invention, the dynamically varying video comprises at least one local ticker, a global ticker and at least one scrolling text ticker. Referring to Figure 3, the commercial detection system has presented some of the selected frames from news and advertisements. After commercial detection the frames which consists commercial frames can be marked. Accordingly, the marked segment of advertisement can be deleted and successive news video segments can be appended for archival purposes for firmer analysis thereof.
Figure: 4(A) shows a news frame from an English news Channel. The regions in the video frame are identified as R1, R2, R3 and R4 each representing a separate ticker band. Among these ticker bands R1 is local video ticker and R2 is a local ticker text bands, R3 is the global ticker text band and R4 is the scrolling ticker text band.
Referring to Figure 4(B), two successive frames of a broadcasted video are shown that dispicts the abrupt change in the regions identified in Figure 4(A). The (b) shows the change in the position of global ticker text band R3 in comparison to (a) which comprises regular regions. Frame (b) is from the advertisement part of the news program in which it can be seen that the global ticker text band is shifted. This shift in the position of the global ticker text band forms the primer of the present ineventino to identify the advertisement as shown in (b).
Referring to Figure 4(C), in the said global tcker text band R3 a region of interest (ROI) is identifier according to shifting position or absence thereof. The ROI provides for the sample of video for fragmented analysis. For detection purposes the further tracking is carried out in the ROI.

Figure 4 (D) has shown the image processing in the ROI. The image processing is initiated with binarization of the ROI followed by edge detection of objects appearing in the ROI during news and advertisements. As shown in the FIG 4 (C), the global ticker text band R3 is absent during adverting and edges of textual geometries are absent therein. This edge detection provides for identification of advertisement over a regular news video.

BEST MODE / EXAMPLE OF WORKING OF THE INVENTION:
The invention is described in the example given below which is provided only to illustrate the invention and therefore should not be construed to limit the scope of the invention.
The analysis of video in various situations may be intended to indentify plurality of generic segments in the composite video which has been broadcasted and are generally useful for archival and data storage management. Particularly, a video signals related to broadcasted news program and sports, forms a composite a composite signals of news/sports video signals (termed herein as "Primary signals") and commercial advertising signals (termed herein as "Secondary signals").
The primary signals comprise of plurality of text ticker band and generally the position of a global ticker band is known a priori. This global ticker band typically comprises text contents related to news, score or other details related to the primary signals. The global ticker band is geometrically dynamic and changed its position corresponding to broadcasting of primary and secondary signals. Thus, during commercial advertising, the global ticker text band disappears or goes down for unhindered and imposing broadcasting of commercial advertisement.
Accordingly, a region of interest (ROI) within the global ticker text is predetermined and video signals within the ROI are fragmented into plurality of frames and clustered into a predefined size of window for differential analysis of the signals. A text is detected within the ROI for differentiating primary signals from the secondary signals. Hence, while commercial advertisements are broadcasted, the global ticker text band is absent, hence ROI detects no text therein. Whereas, while news or sports signals are broadcasted, a global ticker text band is visible in the broadcasted video and text is detected. Hence a change in non text to text and text to non text detection in the ROI results in determining the starting and ending instances associated with both primary and secondary signals.
Thus, according to the detected instances of the commercial advertisements, the portion of advertisements can be deleted from the composite broadcasted video signals and unitary or generic contents of single category can be stored or achieved in the databases for further

analysis. Interestingly, taking into consideration that the quantum of broadcast is 24 by 7 throughout year, huge contents are generated, deleting unwanted secondary signals provides for optimization of archival and databases systems.
The methodology and techniques described with respect to the exemplary embodiments can be performed using a machine or other computing device within which a set of instructions, when executed, may cause the machine to perform any one or more of the methodologies discussed above. In some embodiments, the machine operates as a standalone device. In some embodiments, the machine may be connected (e.g., using a network) to other machines. In a networked deployment, the machine may operate in the capacity of a server or a client user machine in a server-client user network environment, or as a peer machine in a peer-to-peer (or distributed) network environment. The machine may comprise a server computer, a client user computer, a personal computer (PC), a tablet PC, a laptop computer, a desktop computer, a control system, a network router, switch or bridge, or any machine capable of executing a set of instructions (sequential or otherwise) that specify actions to be taken by that machine. Further, while a single machine is illustrated, the term "machine" shall also be taken to include any collection of machines that individually or jointly execute a set (or multiple sets) of instructions to perform any one or more of the methodologies discussed herein.
The machine may include a processor (e.g., a central processing unit (CPU), a graphics processing unit (GPU, or both), a main memory and a static memory, which communicate with each other via a bus. The machine may further include a video display unit (e.g., a liquid crystal displays (LCD), a flat panel, a solid state display, or a cathode ray tube (CRT)). The machine may include an input device (e.g., a keyboard) or touch-sensitive screen, a cursor control device (e.g., a mouse), a disk drive unit, a signal generation device (e.g., a speaker or remote control) and a network interface device.
The disk drive unit may include a machine-readable medium on which is stored one or more sets of instructions (e.g., software) embodying any one or more of the methodologies or functions described herein, including those methods illustrated above. The instructions may also reside, completely or at least partially, within the main memory, the static memory, and/or within the processor during execution thereof by the machine. The main memory and the

processor also may constitute machine-readable media.
Dedicated hardware implementations including, but not limited to, application specific integrated circuits, programmable logic arrays and other hardware devices can likewise be constructed to implement the methods described herein. Applications that may include the apparatus and systems of various embodiments broadly include a variety of electronic and computer systems. Some embodiments implement functions in two or more specific interconnected hardware modules or devices with related control and data signals communicated between and through the modules, or as portions of an application-specific integrated circuit. Thus, the example system is applicable to software, firmware, and hardware implementations.
In accordance with various embodiments of the present disclosure, the methods described herein are intended for operation as software programs running on a computer processor. Furthermore, software implementations can include, but not limited to, distributed processing or component/object distributed processing, parallel processing, or virtual machine processing can also be constructed to implement the methods described herein.
The present disclosure contemplates a machine readable medium containing instructions, or that which receives and executes instructions from a propagated signal so that a device connected to a network environment can send or receive voice, video or data, and to communicate over the network using the instructions. The instructions may further be transmitted or received over a network via the network interface device.
While the machine-readable medium can be a single medium, the term "machine-readable medium" should be taken to include a single medium or multiple media (e.g., a centralized or distributed database, and/or associated caches and servers) that store the one or more sets of instructions. The term "machine-readable medium" shall also be taken to include any medium that is capable of storing, encoding or carrying a set of instructions for execution by the machine and that cause the machine to perform any one or more of the methodologies of the present disclosure,

The term "machine-readable medium" shall accordingly be taken to include, but not be limited to: tangible media; solid-state memories such as a memory card or other package that houses one or more read-only (non-volatile) memories, random access memories, or other re-writable (volatile) memories; magneto-optical or optical medium such as a disk or tape; non-transitory mediums or other self-contained information archive or set of archives is considered a distribution medium equivalent to a tangible storage medium. Accordingly, the disclosure is considered to include any one or more of a machine-readable medium or a distribution medium, as listed herein and including art-recognized equivalents and successor media, in which the software implementations herein are stored.
The illustrations of arrangements described herein are intended to provide a general understanding of the structure of various embodiments, and they are not intended to serve as a complete description of all the elements and features of apparatus and systems that might make use of the structures described herein. Many other arrangements will be apparent to those of skill in the art upon reviewing the above description. Other arrangements may be utilized and derived therefrom, such that structural and logical substitutions and changes may be made without departing from the scope of this disclosure. Figures are also merely representational and may not be drawn to scale. Certain proportions thereof may be exaggerated, while others may be minimized. Accordingly, the specification and drawings are to be regarded in an illustrative rather than a restrictive sense.
The preceding description has been presented with reference to various embodiments. Persons skilled in the art and technology to which this application pertains will appreciate that alterations and changes in the described structures and methods of operation can be practiced without meaningfully departing from the principle, spirit and scope.

CLAIMS:
1. A method for analysis of video signals, the video signals comprised of a plurality sets of primary and secondary signals bundled in a broadcast stream, at least one set of both primary and secondary signals being broadcasted intermittently successive to each other, the method comprising of:
identifying the geometric coordinates of a global ticker text band in the video
signals;
identifying a region of interest (ROI) in the said global ticker text band;
extracting plurality of frames from the identified region of interest;
determining a sliding window size corresponding to a predetermined number of
frames within the said region of interest;
binarizing the region of interest to generate a monochrome image therefrom;
detecting edges in the said monochrome image;
identifying geometrical orientation of the detected edges associated with the said
edged monochrome image in each frame thereby differentiating at least one textual
signal from at least one video signal;
calculating an edge value associated with each frame within the ROI encountered
by the sliding window, wherein the edge value corresponds to the number of
occurrences of horizontal edges;
determining the presence of text in at least one frame based on the edge value above
a predetermined threshold value, wherein continuously applying an edge detection
means to predetermined number of frames in a region of interest of the broadcasted
video signals;
recording an instance and location of starting of secondary signals corresponding to
occurrence of at least one (or predetermined) non text frame detected within a
sliding window of the broadcasted video signals;
recording an instance and location of ending of secondary signals corresponding to
occurrence of at least one (or predetermined) text frame detected within a sliding
window of the broadcasted video signals; and
recording a period corresponding to non text and text frames for differentiating the
set of secondary signals from the set of primary signals.

2. A method as claimed in Claim 1. wherein the primary signals correspond to the primary video signals broadcasted for viewers' consumption and the secondary signals corresponds to commercial contents broadcasted along with the primary signals.
3. A method as claimed in Claim 1, wherein the features in the monochrome images are extracted and intensity of each feature is measured using a Sobel operator for detection of intensity of horizontal edges in the images.
4. A method as claimed in Claim 1, wherein the region of interest in the said monochrome image is predefined in a manner to substantially detect textual geometric variations in least possible portion of the global ticker text band.
5. A method as claimed in Claim 1, wherein the presence of the text is detected by calculating a K-means clustering value which corresponds to number of horizontal edges detected in the region of interest.
6. A method as claimed in Claim 1, wherein the instance of starting of the secondary signals correspond to occurrence of one or more non text frames detected is an indication of "Ad Start".
7. A method as claimed in Claim 1, wherein the instance of ending of the secondary signals correspond to occurrence of one or more text frames detected is an indication of "Ad End".
8. A system for video signals analysis, the video signals comprised of a plurality sets of primary and secondary signals bundled in a broadcast stream, at least one set of both primary and secondary signals being broadcasted intermittently successive to each other, the system comprising of:
a plurality sets of primary video signals comprising of a plurality of dynamically varying video and text signals including a global ticker text band; a plurality sets of secondary signals comprising of a plurality of dynamically varying video signals;

a server configured to intermittently feeding at least one set of secondary signals
into the broadcast stream;
a signal position identification means adapted to instantaneously record an instance
of change in coordinates of at least one primary signal;
a region of interest module adapted to define the boundary of the global ticker text
band for geometric variation analysis of contents thereof;
a video fragmentation means adapted to continuously disintegrate the video signals
corresponding to the identified region of interest into a plurality of differential
frames:
an edge detection means configured to evaluate edge value associated with
occurrence of text geometry in at least one frame of at least one signal in the region
of interest; and
a sliding window means configured to continuously apply the edge detection means
to predetermined number of frames in a region of interest of the global ticker text
band and for evaluated edge value within the window being above a predefined
threshold value confirming presence of text in at least one frame of the frame
window, detecting the instances of occurrence of non text and text frames,
differentiating the set of secondary signals from the set of the primary signals.
9. A system as claimed in Claim 8. wherein dynamically varying video comprises at least one local ticker, a scrolling text ticker and the global ticker text band.
10. A system as claimed in Claim 8, the features in the monochrome images are extracted and intensity of each feature is measured using a Sobel operator for detection of intensity of horizontal edges in the images.
11. A system as claimed in Claim 8, wherein the presence of the text is detected by calculating a K-means clustering value which corresponds to number of horizontal edges detected in the region of interest.

12. A system as claimed in Claim 8, wherein the instance of starting of the secondary signals correspond to occurrence of one or more non text frames detected is an indication of "Ad Start".
13. A system as claimed in Claim 8, wherein the instance of ending of the secondary signals correspond to occurrence of one or more text frames detected is an indication of "Ad End".
14. A system as claimed in Claim 8, wherein at least one set of secondary signals wherein the edge detection means is configured to concatenate plurality of disparate commercial video blocks detected with a presence of text in the region of interest encountered by the sliding window frame within the gap of predetermined minimum number of frames.