Identifying stale background pixels in a video analysis system
View Patent ↗Embodiments of the present invention provide a method and a module for identifying a background of a scene depicted in an acquired stream of video frames that may be used by a video-analysis system. For each pixel or block of pixels in an acquired video frame a comparison measure is determined. The comparison measure depends on difference of color values exhibited in the acquired video frame and in a background image respectively by the pixel or block of pixels and a corresponding pixel and block of pixels in the background image. To determine the comparison measure, the resulting difference is considered in relation to a range of possible color values. If the comparison measure is above a dynamically adjusted threshold, the pixel or the block of pixels is classified as a part of the background of the scene.
1. A method for identifying stale background pixels in a background model of scene depicted in a sequence of video frames, the method comprising:
receiving a raw video frame of the sequence of video frames;
classifying a first set of one or more pixels of the raw video frame as depicting a portion of a background of the scene;
classifying a second set of one or more pixels of the raw video frame as depicting a portion of a foreground of the scene;
for at least a first pixel classified as depicting a portion of the foreground of the scene:
selecting a background pixel in the background model corresponding to the first pixel and a plurality of pixels in the background model neighboring the background pixel,
determining a plurality of changes between color-characteristic values of the plurality of pixels in the background model and respective color-characteristic values of a corresponding plurality of pixels in a video frame previously used to update the color-characteristic values of the background pixel, and
upon determining a difference between the change and the plurality of changes is within a specified threshold, classifying the background pixel in the background model as a stale background pixel; and
updating the color-characteristic values of the background pixel in the background model using the color-characteristic values of the first pixel.
2. The method of claim 1 , further comprising, upon determining the plurality of changes is within the specified threshold, re-classifying the first pixel as depicting a portion of a background of the scene.
3. The method of claim 1 , wherein color-characteristics are selected from a group consisting of a Red (R) color-channel, a Green (G) color-channel, and a Blue (B) color-channel.
4. The method of claim 1 , wherein the background model includes a background image composed from a plurality of pixel color values, and wherein the background images depicts static elements of the scene depicted in the sequence of video frames.
5. The method of claim 1 , wherein the operation further comprises updating the background model using the first set of one or more pixels.
6. The method of claim 1 , wherein the background model includes a short-term background model and a long-term background model.
7. A computer-readable storage medium containing a program which, when executed by a processor, performs an operation for identifying stale background pixels in a background model of scene depicted in a sequence of video frames, the operation comprising:
receiving a raw video frame of the sequence of video frames;
classifying a first set of one or more pixels of the raw video frame as depicting a portion of a background of the scene;
classifying a second set of one or more pixels of the raw video frame as depicting a portion of a foreground of the scene;
for at least a first pixel classified as depicting a portion of the foreground of the scene:
selecting a background pixel in the background model corresponding to the first pixel and a plurality of pixels in the background model neighboring the background pixel,
determining a plurality of changes between color-characteristic values of the plurality of pixels in the background model and respective color-characteristic values of a corresponding plurality of pixels in a video frame previously used to update the color-characteristic values of the background pixel, and
upon determining a difference between the change and the plurality of changes is within a specified threshold, classifying the background pixel in the background model as a stale background pixel; and
updating the color-characteristic values of the background pixel in the background model using the color-characteristic values of the first pixel.
8. The computer-readable storage medium of claim 6 , wherein the operation further comprises, upon determining the plurality of changes is within the specified threshold, re-classifying the first pixel as depicting a portion of a background of the scene.
9. The computer-readable storage medium of claim 6 , wherein color-characteristics are selected from a group consisting of a Red (R) color-channel, a Green (G) color-channel, and a Blue (B) color-channel.
10. The computer-readable storage medium of claim 6 , wherein the background model includes a background image composed from a plurality of pixel color values, and wherein the background images depicts static elements of the scene depicted in the sequence of video frames.
11. The computer-readable storage medium of claim 6 , wherein the operation further comprises updating the background model using the first set of one or more pixels.
12. The computer-readable storage medium of claim 6 , wherein the background model includes a short-term background model and a long-term background model.
13. A system, comprising:
a video input source configured to provide a sequence of video frames, each depicting a scene;
a processor; and
a memory containing a video analysis application, which when executed by the processor is configured to perform an operation for identifying stale background pixels in a background model of scene depicted in a sequence of video frames, the operation comprising:
receiving a raw video frame of the sequence of video frames,
classifying a first set of one or more pixels of the raw video frame as depicting a portion of a background of the scene,
classifying a second set of one or more pixels of the raw video frame as depicting a portion of a foreground of the scene, and
for at least a first pixel classified as depicting a portion of the foreground of the scene:
selecting a background pixel in the background model corresponding to the first pixel and a plurality of pixels in the background model neighboring the background pixel,
determining a plurality of changes between color-characteristic values of the plurality of pixels in the background model and respective color-characteristic values of a corresponding plurality of pixels in a video frame previously used to update the color-characteristic values of the background pixel, and
upon determining a difference between the change and the plurality of changes is within a specified threshold, classifying the background pixel in the background model as a stale background pixel; and
updating the color-characteristic values of the background pixel in the background model using the color-characteristic values of the first pixel.
14. The system of claim 13 , further comprising, upon determining the plurality of changes is within the specified threshold, re-classifying the first pixel as depicting a portion of a background of the scene.
15. The system of claim 13 , wherein color-characteristics are selected from a group consisting of a Red (R) color-channel, a Green (G) color-channel, and a Blue (B) color-channel.
16. The system of claim 13 , wherein the background model includes a background image composed from a plurality of pixel color values, and wherein the background images depicts static elements of the scene depicted in the sequence of video frames.
17. The system of claim 13 , wherein the operation further comprises updating the background model using the first set of one or more pixels.
18. The system of claim 13 , wherein the background model includes a short-term background model and a long-term background model.