Object detection based on image pixels

• US 9,412,269 B2
• Filed: 10/16/2015
• Issued: 08/09/2016
• Est. Priority Date: 11/15/2012
• Status: Active Grant

• Current Assignee: Motorola Solutions, Inc.
• Original Assignee: Avigilon Analytics Corporation
• Inventors: Saptharishi, Mahesh, Marman, Douglas H.
• Primary Examiner(s): Harold, Jefferey
• Assistant Examiner(s): Rahman, Mustafizur
• Application Number:
  US14/885,479
• Publication Number:
  US 20160044286A1
• Time in Patent Office: 298 Days
• Field of Search: 348/135, 348/143, 348/142, 348/149, 348/153, 348/154, 348/156, 348/158, 348/169, 348/222.1, 348/565, 382/103, 382/107, 382/154, 382/166, 382/282, 715/700, 715/764, 345/419, 345/652
• US Class Current: 1/1
• CPC Class Codes: G06T 15/08 Volume rendering
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 2207/30241 Trajectory
• CPC Class Codes: G06T 2207/30242 Counting objects in image
• CPC Class Codes: G06T 2207/30252 Vehicle exterior; Vicinity of vehicle
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G06T 7/215 Motion-based segmentation
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G06V 20/44 Event detection
• CPC Class Codes: G06V 20/49 Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes
• 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 20/54 of traffic, e.g. cars on the road, trains or boats
• CPC Class Codes: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition
• CPC Class Codes: G08B 13/19613 Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion
• CPC Class Codes: G08B 13/2491 Intrusion detection systems, i.e. where the body of an intruder causes the interference with the electromagnetic field
• CPC Class Codes: G08G 1/0133 for classifying traffic situation
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources
• CPC Class Codes: H04N 7/183 : for receiving images from a single remote source
• CPC Class Codes: H04N 7/188 : Capturing isolated or intermittent images triggered by the occurrence of a predetermined event, e.g. an object reaching a predetermined position signal generation from motion picture films H04N5/253

Abstract

Video analytics is used to track an object of interest represented in video data representing the field of view of a scene observed by a video camera. A multidimensional virtual beam is used to detect whether the tracked object of interest is continually present in a detection zone within the field view of the scene. An occurrence of an event is signaled when the tracked object of interest is continually present in the detection zone during a period beginning when the tracked object of interest enters the detection zone and ending when the tracked object of interest leaves the detection zone through the opposite side, after having completely crossed through the detection zone. Use of a virtual beam detection zone reduces false alarms as compared to the numbers of incidences of false alarms of traditional detection methods, while adding several features and benefits.

Claims

1. A method of monitoring whether an object of interest is passing through an area within a field of view of a scene observed by a video camera, the method comprising:

receiving video data representing the field of view of the scene observed by the video camera;

using video analytics to track an object of interest represented in the video data, the object of interest corresponding to image pixels in the video data;

configuring a detection zone within the area, the detection zone including an entrance side and an exit side positioned opposite to each other and thereby defining opposite sides of the detection zone; and

signaling an occurrence of an event when the tracked object of interest is continually present in the detection zone during a period beginning when the tracked object of interest enters the detection zone on the entrance side and ending when the tracked object of interest leaves the detection zone from the exit side, wherein the video analytics determines whether the object of interest enters or leaves the detection zone by determining whether an image pixel representing the object of interest is present in the detection zone.

2. The method of claim 1, in which the detection zone corresponds to a volume superimposed on a two-dimensional image represented by the video data.

3. The method of claim 1, in which the detection zone corresponds to a volume of space oriented in a three-dimensional representation of the scene observed by the video camera.

4. The method of claim 1, in which the signaling of an occurrence of an event is performed for an object of interest exhibiting a specified physical characteristic.

5. The method of claim 4, in which the physical characteristic comprises at least one of a color of the object of interest, shape of the object of interest, or a size of the object of interest.

6. The method of claim 4, in which the physical characteristic is a speed of motion of the object of interest.

7. The method of claim 4, in which the physical characteristic is a height of the object of interest.

8. The method of claim 4, in which the detection zone corresponds to a volume of space and thereby defines a virtual beam detection zone that is superimposed on an image represented by the video data, and in which the physical characteristic is a height of the object of interest within the virtual beam detection zone.

9. The method of claim 4, in which the detection zone corresponds to a volume of space and thereby defines a virtual beam detection zone that is superimposed on an image represented by the video data, and in which the height of the object of interest is capable of being configured by a height value set by a user.

10. The method of claim 1, in which the object of interest is of a specified type.

11. The method of claim 1, in which the object of interest is one of a number of objects of interest, and in which the signaling of an occurrence of an event is performed upon determination of a specified number of tracked objects of interest.

12. The method of claim 1, further comprising monitoring passage of time, and in which the signaling of an occurrence of an event is performed after detection of a specified period of time elapsed from a time the object of interest entered through the entrance side and left through the exit side of the detection zone.

13. The method of claim 1, further comprising monitoring an output signal of an external sensor, and in which the signaling of an occurrence of an event is performed upon indication of a predetermined value of the output signal of the external sensor.

14. The method of claim 1, in which the object of interest is one of multiple objects of interest, and further comprising performing a count of the multiple objects of interest.

15. The method of claim 14, in which the multiple objects of interest are people present in a retail establishment, and further comprising using the count of people to monitor retail establishment customer traffic patterns.

16. The method of claim 14, in which the multiple objects of interest are vehicles present in a specified region of vehicle travel, and further comprising using the count of vehicles to monitor vehicle traffic patterns in the specified region.

17. The method of claim 1, in which the tracked object of interest is moving within the field of view of the scene observed by the video camera, and further comprising assigning the detection zone to follow the tracked object of interest as it moves.

18. The method of claim 17, in which the scene observed by the video camera is a vehicle parking site and the tracked object of interest is a vehicle.

19. The method of claim 17, in which the scene observed by the video camera is a hospital facility and the tracked object of interest is a person or hospital equipment.

20. The method of claim 1, in which the object of interest is of a specified type, and further comprising automatically assigning the detection zone to encompass the specified object of interest when it appears in the field of view of the scene observed by the video camera.

21. The method of claim 1, wherein the image pixels include an image pixel centrally located within the object of interest, and wherein the video analytics determines whether the object of interest enters or leaves the detection zone by determining whether the centrally located pixel representing the object of interest is present in the detection zone.

22. The method of claim 1, in which one of the entrance and exit sides of the detection zone is configured to be a certain distance from a ground reference to enable detection of a particular portion of the object of interest.

23. The method of claim 1, further comprising determining a direction the tracked object of interest traveled through the detection zone by determining which sides of the detection zone the tracked object of interest entered and exited.

24. The method of claim 1, wherein a detection sensitivity of determining whether the object of interest enters or leaves the detection zone is adjusted based on a number of the image pixels associated with the object of interest.

25. The method of claim 1, wherein the detection zone corresponds to an area of space defining a virtual beam detection zone that is superimposed on an image represented by the video data, wherein the virtual beam is defined by pixel coordinates.

26. The method of claim 1, wherein the object of interest corresponds to one or more image pixels contained within a bounding box, and wherein the video analytics determines whether the object of interest enters or leaves the detection zone by determining whether any edge of the bounding box is present in the detection zone.

27. A method of monitoring whether an object of interest is passing through an area within a field of view of a scene observed by a video camera, the method comprising:

receiving video data representing the field of view of the scene observed by the video camera;

using video analytics to track an object of interest represented in the video data, the object of interest corresponding to one or more image pixels contained within a bounding box;

configuring a detection zone within the area, the detection zone including an entrance side and an exit side positioned opposite to each other and thereby defining opposite sides of the detection zone; and

signaling an occurrence of an event when the tracked object of interest is continually present in the detection zone during a period beginning when the tracked object of interest enters the detection zone on the entrance side and ending when the tracked object of interest leaves the detection zone from the exit side, and in which the video analytics determines whether the object of interest enters or leaves the detection zone by determining whether any edge of the bounding box is present in the detection zone.

28. The method of claim 27, wherein the bounding box is defined by a set of bounding box coordinates.

29. The method of claim 28, wherein the bounding box coordinates each comprise a set of two or three pixel coordinates.

30. The method of claim 27, further comprising associating the bounding box with the tracked object of interest such that the bounding box moves as the tracked object of interest moves.

31. The method of claim 30, further comprising filtering the bounding box coordinates through a smoothing function between frames of the video data.

32. The method of claim 27, wherein the tracked object of interest is identified as moving within the field of view of the scene observed by the video camera based on blob detection.

33. A method of monitoring whether an object of interest is passing through a detection zone within a field of view of a scene observed by a video camera, the detection zone having opposite sides, the method comprising:

receiving video data representing the field of view of the scene observed by the video camera;

using video analytics to track an object of interest represented in the video data

configuring a detection zone, the detection zone including an entrance side and an exit side positioned opposite to each other and thereby defining opposite sides of the detection zone, wherein at least one of the entrance side or exit side of the detection zone is configured to be a certain distance from a ground reference to enable detection of a particular portion of the object of interest; and

signaling an occurrence of an event when the tracked object of interest is continually present in the detection zone during a period beginning when the tracked object of interest enters the detection zone on the entrance side and ending when the tracked object of interest leaves the detection zone from the exit side.

34. The method of claim 33, wherein the particular portion of the object of interest comprises an upper or lower part of a human body.

35. The method of claim 33, wherein the certain distance is selected to detect an adult and not a child below a threshold height.

36. The method of claim 33, wherein the certain distance is selected to detect a child below a threshold height and not an adult.

37. The method of claim 36, wherein upon detection of the child, the method further comprising:

identifying the adult in close proximity to the child; and

signaling the occurrence of the event if the child leaves the detection zone without the adult.