Video object classification with object size calibration

• US 10,417,493 B2
• Filed: 11/07/2018
• Issued: 09/17/2019
• Est. Priority Date: 03/03/2008
• Status: Active Grant

• Current Assignee: Motorola Solutions, Inc.
• Original Assignee: Avigilon Analytics Corporation (Motorola Solutions, Inc.)
• Inventors: Saptharishi, Mahesh, Lisin, Dimitri A., Lipchin, Aleksey, Reyzin, Igor
• Primary Examiner(s): Tran, Nhan T
• Application Number:
  US16/183,669
• Publication Number:
  US 20190073530A1
• Time in Patent Office: 314 Days
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 2218/12 Classification; Matching
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30232 Surveillance
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06T 7/73 using feature-based methods
• CPC Class Codes: G06V 20/41 Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items segmenting video sequences G06V20/49
• 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: H04N 23/611 where the recognised objects include parts of the human body
• CPC Class Codes: H04N 23/63 by using electronic viewfinders
• CPC Class Codes: H04N 23/631 Graphical user interfaces [GUI] specially adapted for controlling image capture or setting capture parameters
• CPC Class Codes: H04N 23/66 Remote control of cameras or camera parts, e.g. by remote control devices
• CPC Class Codes: H04N 23/661 Transmitting camera control signals through networks, e.g. control via the Internet
• 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

Abstract

A camera system comprises an image capturing device, and connected to it are an object classification module and a calibration module. The object classification module is operable to determine whether or not an object in an image is a member of an object class, and the calibration module is operable to estimate representative sizes of the object. The object classification module may determine a confidence parameter that is used by the calibration module, or conversely, the calibration module may produce a size that is used by the classification module.

Claims

1. A method of improving performance of video analytics for a camera system in response to a detection preference of a system user, comprising:

receiving image data representing multiple images of a scene of a field of view of the camera system, the multiple images including representations of multiple objects, a first set of the multiple objects having members of an object class, and a second set of the multiple objects not having members of the object class;

using video analytics implemented with a general classifier that performs general classifier steps in analyzing the received image data to produce a general classification determination classifying the multiple objects as either members or non-members of the object class;

generating mistake metadata in response to acknowledgement by the system user that the general classification determination resulted in a mistaken classification determination based on the detection preference of the system user; and

improving video analytics performance based on the mistake metadata by performing a specialization step in addition to the general classification steps performed, the specialization step producing for presentation to the system user a specialized classification determination of the multiple objects classified by the general classifier so as to reduce a number of future mistaken classification determinations.

2. The method according to claim 1, wherein classifying the multiple objects generates a confidence parameter for each object among the multiple objects, the confidence level representing a measure of confidence that the each object among the multiple objects is a member of the object class.

3. The method according to claim 2 wherein the confidence level is used to determine whether to include a classified object in the multiple objects for presentation to the system user.

4. The method according to claim 1, wherein the general classifier is operable to classify an object as a human or non-human.

5. The method according to claim 1, wherein the general classifier is operable to classify an object as a vehicle or non-vehicle.

6. A camera system comprising:

video analytics for processing image data representing multiple images of a scene of a field of view of the camera system, the multiple images including representations of multiple objects, a first set of the multiple objects having members of an object class, and a second set of the multiple objects not having members of the object class, the video analytics comprising;

a general classifier for performing general classifier steps in analyzing the received image data to produce a general classification determination classifying the multiple objects as either members or non-members of the object class;

wherein the video analytics is operable to generate mistake metadata in response to acknowledgement by the system user that the general classification determination resulted in a mistaken classification determination based on the detection preference of the system user; and

wherein the video analytics is further operable to improve performance based on the mistake metadata by performing a specialization step in addition to the general classification steps performed, the specialization step producing for presentation to the system user a specialized classification determination of the multiple objects classified by the general classifier so as to reduce a number of future mistaken classification determinations.

7. The method according to claim 6, wherein classifying the multiple objects generates a confidence parameter for each object among the multiple objects, the confidence level representing a measure of confidence that the each object among the multiple objects is a member of the object class.

8. The method according to claim 7 wherein the confidence level is used to determine whether to include a classified object in the multiple objects for presentation to the system user.

9. The method according to claim 6, wherein the general classifier is operable to classify an object as a human or non-human.

10. The method according to claim 6, wherein the general classifier is operable to classify an object as a vehicle or non-vehicle.