Method and system for detecting cuts in a video signal

• US 5,767,923 A
• Filed: 06/07/1996
• Issued: 06/16/1998
• Est. Priority Date: 06/07/1996
• Status: Expired due to Term

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Electronic Data Systems Corporation (Perspecta, Inc.)
• Inventors: Coleman, Jr., E. North
• Primary Examiner(s): Kostak, Victor R.
• Application Number:
  US08/660,354
• Time in Patent Office: 739 Days
• Field of Search: 348/699-701, 348/558, 348/604, 348/722, 382/107, 382/190, 382/195, 382/197, 382/209, 382/286
• US Class Current: 348/700
• CPC Class Codes: G11B 2220/20 Disc-shaped record carriers
• CPC Class Codes: G11B 27/28 by using information signals recorded by the same method as the main recording G11B27/22 takes precedence
• CPC Class Codes: H04N 19/142 Detection of scene cut or scene change
• CPC Class Codes: H04N 19/179 the unit being a scene or a shot
• CPC Class Codes: H04N 19/85 using pre-processing or post-processing specially adapted for video compression
• CPC Class Codes: H04N 19/87 involving scene cut or scene change detection in combination with video compression
• CPC Class Codes: H04N 5/147 Scene change detection

Abstract

A method and apparatus for detecting cuts in a digital video signal made up of a series of video images is disclosed. The method comprises segmenting a plurality of the video images into a number of cells, each comprised of a number of pixels having a pixel intensity value. Then, a plurality of cell contrast vectors each associated with one of the segmented video images is generated. An element of the cell contrast vector comprises the standard deviation of the pixel intensity values for the pixels in a particular cell. A cut detect signal is generated for a video image in response to the cell contrast vector associated with that image.

Claims

1. A method of detecting a cut in a digital video signal made up of a series of video images, comprising:

segmenting a plurality of the video images, each of the plurality of video images segmented into a first number of cells, each cell comprising a second number of pixels, a pixel having a pixel intensity value representing the intensity of the pixel;

generating a plurality of cell contrast vectors each associated with one of the segmented video images, each element of the cell contrast vector associated with one of the cells and comprising the standard deviation of the pixel intensity values for the pixels in that cell;

generating a cut-detect signal for a video image in response to the cell contrast vector associated with that image.

2. The method of claim 1, wherein each video image comprises a field of a video signal.

3. The method of claim 1, wherein each video image comprises a frame of a video signal.

4. The method of claim 1, further comprising:

computing a plurality of inter-image similarity values each associated with one of the segmented video images, the inter-image similarity value for a video image comprising the cosine of the angle between the cell contrast vector for that video image and the cell contrast vector for another video image; and

wherein the step of generating a cut-detect signal further comprises generating a cut-detect signal for a video image in response to the inter-image similarity value associated with that image.

5. The method of claim 4, further comprising:

filtering the inter-image similarity values using a spike filter to produce a filtered spike size value for each of the segmented video images; and

wherein the step of generating a cut-detect signal further comprises generating a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a first threshold.

6. The method of claim 4, wherein the inter-image similarity signal for a video image further comprises the cosine of the angle between the cell contrast vector for that video image and the cell contrast vector for the immediately previous video image of the digital video signal.

7. The method of claim 5, further comprising:

generating a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a second threshold and the ratio of the filtered spike size value for that video image to the difference between one and the inter-image similarity value for that video image exceeds a third threshold.

8. The method of claim 5, further comprising:

generating a plurality of contrast change vectors each associated with one of the segmented video images, the contrast change vector for a video image comprising the cell contrast vector for that video image minus the cell contrast vector for another video image;

computing a contrast difference value for each of the segmented video images, the contrast difference value for a video image comprising a count of all elements of the contrast change vector associated with that video image that are greater than a first contrast change value or less than a second contrast change value; and

generating a cut-event signal for a video image if a cut-detect signal has been generated for that video image and the contrast difference value for that video image exceeds a fourth threshold.

9. The method of claim 8, further comprising:

generating a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a second threshold and the ratio of the filtered spike size value for that video image to the difference between one and the inter-image similarity value for that video image exceeds a third threshold.

10. The method of claim 8, further comprising:

generating a cut-detect signal for a video image if the contrast difference value for that video image exceeds a fifth threshold and the difference between the contrast difference value for that video image and the maximum of the contrast difference values for a plurality of other video images exceeds a sixth threshold.

11. The method of claim 8, wherein each video image comprises a field of a video signal;

and

wherein the contrast change vector for a video image comprises the cell contrast vector for that video image minus the cell contrast vector for the same field in the immediately prior frame of the digital video signal.

12. A computerized system for detecting a cut in a digital video signal made up of a series of video images, comprising:

a computer-readable medium; and

a computer program encoded on the computer-readable medium, the computer program further operable to

segment a plurality of the video images, each of the plurality of video images segmented into a first number of cells, each cell comprising a second number of pixels, a pixel having a pixel intensity value representing the intensity of the pixel;

generate a plurality of cell contrast vectors each associated with one of the segmented video images, each element of the cell contrast vector associated with one of the cells and comprising the standard deviation of the pixel intensity values for the pixels in that cell; and

generate a cut-detect signal for a video image in response to the cell contrast vector associated with that image.

13. The computerized system of claim 12 wherein the computer program is further operable to:

compute a plurality of inter-image similarity values each associated with one of the segmented video images, the inter-image similarity value for a video image comprising the cosine of the angle between the cell contrast vector for that video image and the cell contrast vector for another video image; and

wherein the computer program generates a cut-detect signal for a video image in response to the inter-image similarity value associated with that image.

14. The computerized system of claim 13 wherein the computer program is further operable to:

filter the inter-image similarity values using a spike filter to produce a filtered spike size value for each of the segmented video images; and

generate a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a first threshold.

15. The computerized system of claim 14 wherein the computer program is further operable to:

generate a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a second threshold and the ratio of the filtered spike size value for that video image to the difference between one and the inter-image similarity value for that video image exceeds a third threshold.

16. The computerized system of claim 14 wherein the computer program is further operable to:

generate a plurality of contrast change vectors each associated with one of the segmented video images, the contrast change vector for a video image comprising the cell contrast vector for that video image minus the cell contrast vector for another video image;

compute a contrast difference value for each of the segmented video images, the contrast difference value for a video image comprising a count of all elements of the contrast change vector associated with that video image that are greater than a first contrast change value or less than a second contrast change value; and

generate a cut-event signal for a video image if a cut-detect signal has been generated for that video image and the contrast difference value for that video image exceeds a fourth threshold.

17. The computerized system of claim 14 wherein the computer program is further operable to:

generate a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a second threshold and the ratio of the filtered spike size value for that video image to the difference between one and the inter-image similarity value for that video image exceeds a third threshold.

18. The computerized system of claim 14 wherein the computer program is further operable to:

generate a cut-detect signal for a video image if the contrast difference value for that video image exceeds a fifth threshold and the difference between the contrast difference value for that video image and the maximum of the contrast difference values for a plurality of other video images exceeds a sixth threshold.

19. A computerized system for detecting a cut in a digital video signal made up of a series of video images, comprising:

a storage medium;

a processor coupled to the storage medium;

a digital video signal source coupled to the storage medium and providing the digital video signal to the system;

a computer program stored in the storage medium, the computer program operable to run on the processor and process the digital video signal, the computer program further operable to

segment a plurality of the video images, each of the plurality of video images segmented into a first number of cells, each cell comprising a second number of pixels, a pixel having a pixel intensity value representing the intensity of the pixel;

generate a plurality of cell contrast vectors each associated with one of the segmented video images, each element of the cell contrast vector associated with one of the cells and comprising the standard deviation of the pixel intensity values for the pixels in that cell;

generate a cut-detect signal for a video image in response to the cell contrast vector associated with that image.

20. The computerized system of claim 19 wherein the computer program is further operable to:

compute a plurality of inter-image similarity values each associated with one of the segmented video images, the inter-image similarity value for a video image comprising the cosine of the angle between the cell contrast vector for that video image and the cell contrast vector for another video image; and

wherein the computer program generates a cut-detect signal for a video image in response to the inter-image similarity value associated with that image.

21. The computerized system of claim 20 wherein the computer program is further operable to:

filter the inter-image similarity values using a spike filter to produce a filtered spike size value for each of the segmented video images; and

generate a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a first threshold.

22. The computerized system of claim 20 wherein the computer program is further operable to:

generate a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a second threshold and the ratio of the filtered spike size value for that video image to the difference between one and the inter-image similarity value for that video image exceeds a third threshold.

23. The computerized system of claim 20 wherein the computer program is further operable to

generate a plurality of contrast change vectors each associated with one of the segmented video images, the contrast change vector for a video image comprising the cell contrast vector for that video image minus the cell contrast vector for another video image;

compute a contrast difference value for each of the segmented video images, the contrast difference value for a video image comprising a count of all elements of the contrast change vector associated with that video image that are greater than a first contrast change value or less than a second contrast change value; and

generate a cut-event signal for a video image if a cut-detect signal has been generated for that video image and the contrast difference value for that video image exceeds a fourth threshold.

24. A method of detecting a cut in a digital video signal made up of a series of video images, comprising:

segmenting a plurality of the video images, each of the plurality of video images segmented into a first number of cells, each cell comprising a second number of pixels, a pixel having a pixel intensity value representing the intensity of the pixel;

generating a plurality of cell intensity vectors each associated with one of the segmented video images, each element of the cell intensity vector associated with one of the cells and proportional to the average of the pixel intensity values for the pixels in that cell;

generating a cut-detect signal for a video image in response to the cell intensity vector associated with that image.

25. The method of claim 24, further comprising:

computing a plurality of inter-image similarity values each associated with one of the segmented video images, the inter-image similarity value for a video image comprising the cosine of the angle between the cell intensity vector for that video image and the cell intensity vector for another video image; and

wherein the step of generating a cut-detect signal further comprises generating a cut-detect signal for a video image in response to the inter-image similarity value associated with that image.

26. The method of claim 25, further comprising:

filtering the inter-image similarity values using a spike filter to produce a filtered spike size value for each of the segmented video images; and

wherein the step of generating a cut-detect signal further comprises generating a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a first threshold.

27. The method of claim 25, wherein the inter-image similarity value for a video image further comprises the cosine of the angle between the cell intensity vector for that video image and the cell intensity vector for the immediately previous video image of the digital video signal.

28. The method of claim 26, further comprising:

generating a cut-detect signal for a video image if the filtered spike size value for that video image exceeds a second threshold and the ratio of the filtered spike size value for that video image to the difference between one and the inter-image similarity value for that video image exceeds a third threshold.