Apparatus, methods and computer products for video analytics

• US 10,229,322 B2
• Filed: 04/06/2017
• Issued: 03/12/2019
• Est. Priority Date: 04/06/2017
• Status: Active Grant

• Current Assignee: KAMI Vision Incorporated
• Original Assignee: ANTS TECHNOLOGY (HK) LIMITED
• Inventors: Fridental, Ron, Blayvas, Ilya
• Primary Examiner(s): Couso, Jose L
• Application Number:
  US15/480,709
• Publication Number:
  US 20180293442A1
• Time in Patent Office: 705 Days
• CPC Class Codes: G06F 16/5854 using shape and object relationship
• CPC Class Codes: G06F 16/7837 using objects detected or recognised in the video content
• CPC Class Codes: G06F 18/217 Validation; Performance evaluation; Active pattern learning techniques
• CPC Class Codes: G06V 10/776 Validation; Performance evaluation
• 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 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition

Abstract

A computerized method for providing no-configuration object recognition video analytics, the method comprising generating a database stored in computer memory and including labelled images of recognized objects imaged at an initial set of sites including at least one site; using the database as a training set to train a pattern recognition engine; uploading images of additional site/s including at least one additional site which is not a member in the initial set of sites; providing object recognition results for the additional site/s by using the pattern recognition engine on the images as uploaded; verifying the object recognition results; and using a processor for operating the pattern recognition engine on images generated at each the additional site/s including modifying the pattern recognition engine according to and when indicated by the verifying, thereby to provide no-configuration object recognition video analytics at the additional site/s.

Claims

1. A computerized method for simplifying an installation of a video analytics system in an environment, the method comprising:

generating a database of labelled images of recognized objects,

wherein the images were imaged at an initial set of sites and

wherein the set of sites includes at least one site;

using the database as a training set to train a pattern recognition engine;

uploading images of additional one or more sites including at least one additional site which is not a member in the initial set of sites;

providing object recognition results for the additional one or more sites by using the pattern recognition engine on said images as uploaded;

verifying said object recognition results; and

using a processor for operating the pattern recognition engine on images generated at each said additional one or more sites including modifying the pattern recognition engine at the additional one or more sites accordingly and when indicated by said verifying, thereby to provide automatic configuration of video analytics system'"'"'s camera at said additional one or more sites including accommodating unique characteristics of each of the one or more sites,

wherein said pattern recognition engine is trained to identify a type of said environment, and

wherein the method further selects at least one video analytics application that is suitable for the identified environment.

2. A method according to claim 1, wherein said verifying is performed remotely from the additional one or more sites.

3. A method according to claim 1 wherein said verifying comprises presenting said object recognition results to a human and accepting a verifying input provided by the human.

4. A method according to claim 3, wherein said accepting comprises displaying two selectable options to the human, corresponding to “

object” and

“

not object” and

accepting an option that the human selected out of one of the two options.

10. A method according to claim 1, wherein the method further configures at least one selected video analytics application that is suitable for the identified environment.

11. A method according to claim 1, wherein said pattern recognition engine is trained to classify a type of said environment.

12. A method according to claim 1, wherein said providing of object recognition results includes installment and deployment of the one or more cameras for the additional one or more sites.

13. A method according to claim 1, wherein said verifying, of said object recognition results, is performed automatically by a server.

5. A computerized system for simplifying an installation of a video analytics system in an environment, the system comprising:

a database including labelled images of objects recognized at an initial set of sites including at least one site;

a user interface for accepting uploaded images of additional one or more sites including at least one additional site which is not a member in an initial set of sites

wherein the uploaded images are stored in said database; and

a processor configured for;

(i) using said database as a training set to train a pattern recognition engine;

(ii) providing object recognition results for the additional one or more sites by using a pattern recognition engine on said images as uploaded wherein said pattern recognition engine is in data communication with said processor;

(iii) verifying said object recognition results;

(iv) operating the pattern recognition engine on images generated at each said additional one or more sites including modifying the pattern recognition engine accordingly; and

(v) automating configuration of said video analytics system'"'"'s camera at said additional one or more sites including accommodating unique characteristics of each of the one or more sites, based on said verifying,

wherein said pattern recognition engine is trained to identify a type of said environment, and

wherein the processor is further selecting at least one video analytics application that is suitable for the identified environment.

7. A system according to claim 5 and wherein said pattern recognition engine comprises a semantic segmentation engine.

8. A system according to claim 5 and wherein said pattern recognition engine comprises an object detection engine.

9. A system according to claim 5 and wherein said pattern recognition engine comprises an object recognition engine.

6. A computer program product, comprising a non-transitory tangible computer readable medium having computer readable program code embodied therein, said computer readable program code adapted to be executed to implement a computerized method for simplifying an installation of video analytics system in an environment, the method comprising:

generating a database and including labelled images of recognized objects imaged at an initial set of sites including at least one site;

using the database as a training set to train a pattern recognition engine;

uploading images of additional one or more sites including at least one additional site which is not a member in the initial set of sites, wherein the uploaded images are stored in said database;

providing object recognition results for the additional one or more sites by using the pattern recognition engine on said images as uploaded;

verifying said object recognition results; and

using a processor for operating the pattern recognition engine on images generated at each said additional one or more sites including modifying the pattern recognition engine according to and when indicated by said verifying, thereby to automate configuration of video analytics system'"'"'s camera at said additional one or more sites, including accommodating unique characteristics of each of the one or more sites, based on said verifying,

wherein said pattern recognition engine is trained to identify a type of said environment, and

wherein the method further selects at least one video analytics application that is suitable for the identified environment.