Method for rating areas in video frames

• US 8,457,402 B2
• Filed: 11/23/2009
• Issued: 06/04/2013
• Est. Priority Date: 12/10/2008
• Status: Expired due to Fees

• Current Assignee: Artimedia PTE Ltd. (Artivision Technologies Ltd.)
• Original Assignee: Artimedia PTE Ltd. (Artivision Technologies Ltd.)
• Inventors: Miller, Ofer, Segev, Amir
• Primary Examiner(s): Alavi, Amir
• Application Number:
  US13/131,264
• Publication Number:
  US 20120017238A1
• Time in Patent Office: 1,289 Days
• Field of Search: 382/103, 382/164, 382/173, 382/190, 382218-220, 382/224, 348 42- 60, 725/32, 375/240.01, 375/240.08, 375/240.16
• US Class Current: 382/173
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/174 involving the use of two or more images
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: H04H 60/48 for recognising items expressed in broadcast information

Abstract

A method is provided for rating areas in video frames. The method comprises: (a) providing a video signal comprising a plurality of consecutive video frames; (b) selecting a frame from among these frames and indentifying a first plurality of objects comprised in the selected frame; (c) upon comparing the selected frame with other frames, dividing the first plurality of the objects into a second plurality of background objects and a third plurality of foreground objects; (d) dividing objects of the third plurality of foreground objects into one or more connected components, where a connected component comprises group(s) of pixels wherein a change in each of these pixels is associated with a change in each of the remaining pixels of the respective group(s); and (e) associating an interest level with the connected components based upon a pre-determined set of criteria, thereby providing a rating of areas comprised within the video signal.

Claims

1. A method for rating areas comprised in video frames included in a video signal, and comprising:

(a) providing a video signal comprising a plurality of consecutive video frames;

(b) selecting a frame from among said plurality of consecutive video frames, being a selected frame, and identifying a first plurality of objects comprised in said selected frame;

(c) upon comparing said selected frame with one or more other of said plurality of consecutive frames, dividing a plurality of said objects into a second plurality of background objects and a third plurality of foreground objects;

(d) dividing at least one object from among said third plurality of foreground objects into one or more connected components, where the connected component comprises at least one group of pixels and wherein a change in each of the pixels comprised in said at least one group of pixels is associated with a change in each of the remaining pixels of said at least one group of pixels; and

(e) associating an interest level with said one or more connected components based upon a pre-determined set of criteria, thereby providing a rating of areas comprised within said video signal.

2. A method according to claim 1, further comprising a step of inserting one or more visual messages into one or more of said rated areas based upon an interest level associated therewith.

3. A method according to claim 1, further comprising a step dividing at least one object from among said second plurality of background objects into one or more connected components, thereby allowing to rate areas comprised in objects included in said second plurality of background objects.

4. A method according to claim 1, wherein an object that has been classified as a foreground object, retains this classification in each of the proceeding video frames that said object would appear in, for as long as that sequence of frames lasts.

5. A method according to claim 1, wherein an object which belongs to said third plurality of foreground objects is re-classified as a background object provided that no criteria for classifying an object as a foreground object are met for said object for at least a pre-defined number of consecutive video frames.

6. A method according to claim 1, further comprising a step of identifying areas comprised within objects included in said plurality of video frames that their respective rating has been changed along said video signal.

7. A method according to claim 6, wherein said change in the rating is of at least a pre-defined number of levels of interest.

8. A method according to claim 1, further comprising a step of matching at least one object from among said first plurality of objects to one or more pre-defined characterizing classes.

9. A method according to claim 8, further comprising a step of inserting one or more visual messages at one or more areas associated with said at least one object that matches the one or more pre-defined characterizing classes, and wherein said insertion is made in accordance with the respective level of interest of said area and the characterizing class with which said at least one object has been matched.

10. A non-transitory computer readable medium storing a computer program for executing a set of instructions by a computer system comprising one or more computer processors a process for rating areas comprised in video frames which are included in a video signal, the process comprising:

(a) selecting a frame from among a plurality of consecutive video frames comprised in a video signal provided, said frame being a selected frame;

(b) identifying a first plurality of objects comprised in said selected frame;

(c) upon comparing the selected frame with one or more of said plurality of consecutive video frames, deriving a second plurality of background objects and a third plurality of foreground objects;

(d) dividing at least one object from among said third plurality of foreground objects into one or more connected components; and

(e) associating an interest level with said one or more connected components based upon a pre-determined set of criteria, thereby providing a rating for areas comprised within said plurality of video frames.

11. A non-transitory computer readable medium according to claim 10, wherein said process further comprises a step of inserting one or more visual messages into one or more areas comprised within said plurality of video frames while complying with one or more pre-defined rating criteria.

12. A non-transitory computer readable medium according to claim 10, wherein said process further comprises a step of matching at least one object with one or more pre-defined characterizing classes.

13. A non-transitory computer readable medium according to claim 12, wherein said process further comprises a step of inserting one or more visual messages at one or more areas associated with said at least one object in accordance with a respective level of interest of said one or more areas and the characterizing class to which said at least one object has been matched.