Long-term static object detection

• US 9,754,178 B2
• Filed: 08/19/2015
• Issued: 09/05/2017
• Est. Priority Date: 08/27/2014
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Fan, Quanfu, Lu, Zuoxuan, Pankanti, Sharathchandra U.
• Primary Examiner(s): Mistry, Oneal R
• Application Number:
  US14/830,273
• Publication Number:
  US 20160063344A1
• Time in Patent Office: 748 Days
• CPC Class Codes: G06T 2207/30236 Traffic on road, railway or crossing
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06V 10/44 Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 2201/08 Detecting or categorising vehicles

Abstract

Software for static object detection that performs the following operations: (i) detecting an object that is present in at least one image of a set of images, wherein the set of images correspond to a time period; (ii) identifying a set of corner points for the detected object; (iii) tracking the object'"'"'s presence in the set of images over the time period, wherein the object'"'"'s presence is determined by matching the set of images to a template generated based on the identified corner points; and (iv) identifying the object as a static object when an amount of time corresponding to the object'"'"'s presence in the set of images is greater than a predefined threshold.

Claims

1. A method for static object detection, the method comprising:

detecting an object that is present in at least one image of a set of images, wherein the set of images correspond to a time period;

identifying a set of corner points for the detected object;

generating a template based on the identified set of corner points, wherein the template includes a weight assigned to a corner point of the set of corner points;

tracking the object'"'"'s presence in the set of images over the time period, wherein the object'"'"'s presence is determined by matching the set of images to the template;

determining background information, wherein the background information includes an amount of a match between the template and a background of a first image of the set of images, and wherein the background is determined utilizing background subtraction;

updating the template based on the background information, wherein updating the template includes adjusting the weight based on the amount of the match between the background and the generated template; and

identifying the object as a static object when an amount of time corresponding to the object'"'"'s presence in the set of images is greater than a predefined threshold;

wherein at least the tracking is performed by computer software running on computer hardware.

2. The method of claim 1, wherein the set of images includes frames from a video.

3. The method of claim 1, further comprising generating an alert indicating that the static object has been detected.

4. The method of claim 1, further comprising:

identifying additional corner points for the detected object during the tracking; and

updating the template based, at least in part, on the identified additional corner points.

5. The method of claim 1, further comprising utilizing a tree-based representation to manage detection of the object when the object is occluded.

6. The method of claim 1, further comprising:

determining an amount of stability for corner points in the set of corner points, wherein the weight assigned to the corner point of the set of corner points is based, at least in part, on the corner point'"'"'s respective amount of stability.

7. The method of claim 1, wherein the adjusting of the weight includes decreasing the weight when the amount of the match between the template and the background of the first image is greater than or equal to an amount of a match between the template and a frame of the first image, the frame being defined by the identified corner points.

8. The method of claim 1, wherein the first image is matched to the template when the amount of the match between the template and the background of the first image is less than an amount of a match between the template and a frame of the first image, the frame being defined by the identified corner points.

9. The method of claim 8, wherein the first image is matched to the template when a matching score S computed from the following formula is above a predetermined threshold:

10. A computer program product for static object detection, the computer program product comprising a computer readable storage medium having stored thereon:

instructions executable by a device to cause the device to detect an object that is present in at least one image of a set of images, wherein the set of images correspond to a time period;

instructions executable by the device to cause the device to identify a set of corner points for the detected object;

instructions executable by the device to cause the device to generate a template based on the identified set of corner points, wherein the template includes a weight assigned to a corner point of the set of corner points;

instructions executable by the device to cause the device to track the object'"'"'s presence in the set of images over the time period, wherein the object'"'"'s presence is determined by matching the set of images to the template;

instructions executable by the device to cause the device to determine background information, wherein the background information includes an amount of a match between the template and a background of a first image of the set of images, and wherein the background is determined utilizing background subtraction;

instructions executable by the device to cause the device to update the template based on the background information, wherein updating the template includes adjusting the weight based on the amount of the match between the background and the generated template; and

instructions executable by the device to cause the device to identify the object as a static object when an amount of time corresponding to the object'"'"'s presence in the set of images is greater than a predefined threshold.

11. The computer program product of claim 10, wherein the set of images includes frames from a video.

12. The computer program product of claim 10, further comprising instructions executable by the device to cause the device to generate an alert indicating that the static object has been detected.

13. The computer program product of claim 10, further comprising:

instructions executable by the device to cause the device to identify additional corner points for the detected object during the tracking; and

instructions executable by the device to cause the device to update the template based, at least in part, on the identified additional corner points.

14. The computer program product of claim 10, further comprising instructions executable by the device to cause the device to utilize a tree-based representation to manage detection of the object when the object is occluded.

15. The computer program product of claim 10, further comprising:

instructions executable by the device to cause the device to determine an amount of stability for corner points in the set of corner points, wherein the weight assigned to the corner point of the set of corner points is based, at least in part, on the corner point'"'"'s respective amount of stability.

16. A computer system for static object detection, the computer system comprising:

a processor(s) set; and

a computer readable storage medium;

wherein;

the processor set is structured, located, connected and/or programmed to execute instructions stored on the computer readable storage medium; and

the instructions include;

instructions executable by the processor set to cause the system to detect an object that is present in at least one image of a set of images, wherein the set of images correspond to a time period;

instructions executable by the processor set to cause the system to identify a set of corner points for the detected object;

instructions executable by the processor set to cause the system to generate a template based on the identified set of corner points, wherein the template includes a weight assigned to a corner point of the set of corner points;

instructions executable by the processor set to cause the system to track the object'"'"'s presence in the set of images over the time period, wherein the object'"'"'s presence is determined by matching the set of images to the template;

instructions executable by processor set to cause the system to determine background information, wherein the background information includes an amount of a match between the template and a background of a first image of the set of images, and wherein the background is determined utilizing background subtraction;

instructions executable by the processor set to cause the system to update the template based on the background information, wherein updating the template includes adjusting the weight based on the amount of the match between the background and the generated template; and

instructions executable by the processor set to cause the system to identify the object as a static object when an amount of time corresponding to the object'"'"'s presence in the set of images is greater than a predefined threshold.

17. The computer system of claim 16, wherein the set of images includes frames from a video.

18. The computer system of claim 16, further comprising instructions executable by the processor set to cause the system to generate an alert indicating that the static object has been detected.

19. The computer system of claim 16, further comprising:

instructions executable by the processor set to cause the system to identify additional corner points for the detected object during the tracking; and

instructions executable by the processor set to cause the system to update the template based, at least in part, on the identified additional corner points.

20. The computer system of claim 16, further comprising instructions executable by the processor set to cause the system to utilize a tree-based representation to manage detection of the object when the object is occluded.