METHOD AND APPARATUS FOR DETECTING IMPORTANT INFORMATION FROM MOVING PICTURE

• US 20110052061A1
• Filed: 03/24/2010
• Published: 03/03/2011
• Est. Priority Date: 08/25/2009
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd., Soongsil University Research Consortium Techno-Park
• Original Assignee: Samsung Electronics Co. Ltd., Soongsil University Research Consortium Techno-Park
• Inventors: CHUN, Sung-kuk, JUNG, Min-kyu, LEE, Dong-keun, JEONG, Jin-guk, JUNG, Kee-chul
• Granted Patent: US 8,929,656 B2
• Time in Patent Office: Days
• US Class Current: 382/173
• CPC Class Codes: G06T 2207/30221 Sports video; Sports image
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06V 20/40 in video content extracting overlay text G06V20/62; video retrieval G06F16/70; processing of video elementary streams in video servers H04N21/234; processing of video elementary streams in video clients H04N21/44

Abstract

Provided is a method of detecting important information from a moving picture. The method includes: detecting first candidate areas that are presumed to include important information in a plurality of moving picture frames by using stop edge information, which is edge information overlapped at a same position throughout the plurality of moving picture frames, from among edge information in at least two received moving picture frames; determining second candidate areas by performing grouping on the stop edge information according to a position of the stop edge information in the first candidate areas; analyzing the second candidate areas determined in the at least two moving picture frames; and detecting important information areas from each of the at least two moving picture frames based on the analysis.

Claims

1. A method of detecting important information, the method comprising:

detecting first candidate areas that are presumed to include important information in a plurality of moving picture frames by using stop edge information, which is edge information overlapped at a same position throughout the plurality of moving picture frames, from among edge information in at least two received moving picture frames;

determining second candidate areas by performing grouping on the stop edge information according to a position of the stop edge information in the first candidate areas;

analyzing the second candidate areas determined in the at least two moving picture frames; and

detecting important information areas from each of the at least two moving picture frames based on the analysis.

2. The method of claim 1, wherein the detecting of the first candidate areas comprises:

respectively dividing the at least two moving picture frames into a plurality of analyzed areas;

detecting the edge information from the plurality of analyzed areas;

detecting the edge information that is overlapped at the same position throughout the plurality of moving picture frames from among the edge information detected from the plurality of analyzed areas as the stop edge information; and

detecting the first candidate areas, from among the analyzed areas, in which a ratio of areas occupied by the stop edge information is above a predetermined critical value.

3. The method of claim 2, wherein in the detecting of the edge information from the plurality of analyzed areas, the analyzed areas which are presumed to not include important information from among the plurality of analyzed areas based on the positions of the analyzed areas in the moving picture frames are excluded from being analyzed for the edge information.

4. The method of claim 1, wherein the determining of the second candidate areas comprises:

dividing one of the moving picture frames into a plurality of grouping areas;

dividing the stop edge information included in the first candidate areas into a plurality of stop edge information;

allocating each of the divided stop edge information to one of the grouping areas based on a distance between center positions of the grouping areas and the divided stop edge information; and

determining the second candidate areas by repeatedly performing grouping based on a distance between centers of the divided stop edge information included in the grouping areas in each of the grouping areas and the divided stop edge information.

5. The method of claim 1, wherein the analyzing the second candidate areas comprises:

converting the stop edge information included in the second candidate areas to be displayed in a predetermined number of colors;

performing binarization on the stop edge information that is converted in the second candidate areas to be displayed in two colors;

dividing the second candidate areas into a plurality of areas; and

analyzing the binarized stop edge information included in each of the divided areas.

6. The method of claim 5, wherein the analyzing of the binarized stop edge information included in each of the divided areas comprises:

determining whether the binarized stop edge information respectively included in the divided areas of the second candidate areas is a character or a number;

recognizing characters or numbers indicated by one of the stop edge information respectively included the divided areas; and

analyzing the binarized stop edge information respectively included in the divided areas based on the recognized characters or numbers.

7. The method of claim 6, wherein in the detecting of the important information areas, the important information areas are detected based on a number of characters and numbers respectively included in the second candidate areas.

8. The method of claim 6, wherein, when the stop edge information included each of the divided areas are recognized as numbers in the detecting of the important information areas, the important information areas are detected based on initial values of the numbers and whether a time consumed for changing the numbers is above a critical value.

9. The method of claim 1, further comprising reproducing a highlight of a sports game based on a time corresponding to when a number of a score of the sports game, from among the important information included in the important information areas, is changed, after detecting the important information areas from the at least two moving picture frames.

10. The method of claim 1, further comprising transmitting one of the important information areas detected from at least two moving picture frames to a third party in an image form or transmitting the important information included in the important information areas to a third party in a message form using characters and numbers, after detecting the important information areas from at least two moving picture frames.

21. A computer readable recording medium having embodied thereon a computer program which causes a computer to execute the method of claim 1.

11. An apparatus for detecting important information, the apparatus comprising:

a first candidate area detecting unit which detects first candidate areas that are presumed to include important information in a plurality of moving picture frames by using stop edge information, which is edge information overlapped at a same position throughout the plurality of moving picture frames, from among edge information in at least two moving picture frames;

a second candidate area determining unit which determines second candidate areas by performing grouping on the stop edge information according to the position of the stop edge information in the first candidate areas;

an analyzing unit which analyzes the second candidate areas determined in the at least two moving picture frames; and

an important information area detecting unit which detects important information areas from each of the at least two moving picture frames based on the analysis.

12. The apparatus of claim 11, wherein the first candidate area detecting unit respectively divides the at least two moving picture frames into a plurality of analyzed areas, detects the edge information from the plurality of analyzed areas, detects the edge information that is overlapped at a same position throughout the plurality of moving picture frames from among the edge information detected from the plurality of analyzed areas as the stop edge information, and from among the analyzed areas, detects the first candidate areas in which a ratio of areas occupied by the stop edge information is above a predetermined critical value.

13. The apparatus of claim 12, wherein the first candidate area detecting unit does not analyze the analyzed areas from among the plurality of analyzed areas for the edge information which are presumed to not include important information based on positions of the analyzed areas in the moving picture frames.

14. The apparatus of claim 11, wherein the second candidate area determining unit divides one of the moving picture frames into a plurality of grouping areas, divides the stop edge information included in the first candidate areas into a plurality of stop edge information, allocates each of the divided stop edge information to one of the grouping areas based on a distance between center positions of the grouping areas and the dividing stop edge information, and determines the second candidate areas by repeatedly performing grouping based on a distance between centers of the divided edge information included in the grouping areas in each of the grouping areas and the divided stop edge information.

15. The apparatus of claim 11, wherein the analyzing unit converts the stop edge information included in the second candidate areas to be displayed in a predetermined number of colors, performs binarization on the stop edge information that is converted in the second candidate areas to be displayed in two colors, and divides the second candidate areas into a plurality of divided areas, and analyzes the binarized stop edge information included in each of the divided areas.

16. The apparatus of claim 15, wherein the analyzing unit determines whether the binarized stop edge information respectively included in the divided areas of the second candidate areas is a character or number, recognizes characters or numbers indicated by one of the stop edge information respectively included in the divided areas, and analyzes the binarized stop edge information respectively included in the divided areas based on the recognized characters or numbers.

17. The apparatus of claim 16, wherein the important information area detecting unit detects the important information areas based on a number of characters and numbers respectively included in the second candidate areas.

18. The apparatus of claim 16, wherein, when the stop edge information included in each of the divided areas are recognized as numbers, the important information area detecting unit detects the important information areas based on initial values of the numbers and whether a time consumed for changing the numbers is above a critical value.

19. The apparatus of claim 11, further comprising a highlight reproducing unit which reproduces a highlight of sports game based on a time corresponding to when a number of a score of the sports game, from among the important information included in the important information areas, is changed, after the important information area detecting unit the detects the important information areas from the at least two moving picture frames.

20. The apparatus of claim 11, further comprising a transmitter which transmits one of the important information areas detected from at least two moving picture frames to a third party in an image form or the important information included in the important information areas to a third party in a message form using characters and numbers, after the important information area detecting unit detects the important information areas from the at least two moving picture frames.