Multi-camera vehicle identification system

• US 9,953,245 B2
• Filed: 08/19/2016
• Issued: 04/24/2018
• Est. Priority Date: 07/02/2015
• Status: Active Grant

• Current Assignee: AGT International GMBH
• Original Assignee: AGT International GMBH
• Inventors: Palacio, Sebastian, Krauss, Stephan, Hirzel, Jan, Stricker, Didier, Schlattmann, Markus, Hohmann, Sebastian
• Primary Examiner(s): Lu, Tom Y
• Application Number:
  US15/241,463
• Publication Number:
  US 20170004386A1
• Time in Patent Office: 613 Days
• Field of Search: 382103, 382105
• CPC Class Codes: G06F 18/214 Generating training patterns; Bootstrap methods, e.g. bagging or boosting
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 18/2411 based on the proximity to a decision surface, e.g. support vector machines
• CPC Class Codes: G06F 18/256 of results relating to different input data, e.g. multimodal recognition
• CPC Class Codes: G06V 10/811 the classifiers operating on different input data, e.g. multi-modal recognition
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats
• CPC Class Codes: G06V 20/62 Text, e.g. of license plates, overlay texts or captions on TV images
• CPC Class Codes: G06V 2201/08 Detecting or categorising vehicles
• CPC Class Codes: G08G 1/015 with provision for distinguishing between two or more types of vehicles, e.g. between motor-cars and cycles
• CPC Class Codes: G08G 1/0175 by photographing vehicles, e.g. when violating traffic rules
• CPC Class Codes: G08G 1/04 using optical or ultrasonic detectors
• CPC Class Codes: H04N 23/90 Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

A method and system of identifying in an image captured by a second camera a target object captured by a first camera is disclosed. The method comprises using image captured by the 1^st camera for generating a first set of objects comprising the target object and other objects; classifying each object to generate a reference group of attribute values characterizing the objects; using image captured by the 2^nd camera for generating second sets of objects; classifying each object in each second set to generate, for each second set, a corresponding group of attribute values characterizing the objects in the corresponding second set; selecting the second set of objects corresponding to the group of attribute values best matching the reference group of attribute values; and identifying the target object in the selected second set of objects in accordance with a position of the target object in the first set of objects.

Claims

1. A method of identifying in an image captured by a second camera a target object captured by a first camera, the method implemented on a processor and comprising, by the processor:

using an image captured by the 1^st camera for generating a first set of n objects comprising the target object and (n−

1) other objects;

classifying each object in the first set to generate a reference group of attribute values characterizing the n objects;

using an image captured by the 2^nd camera for generating a plurality of second sets of objects, each second set of objects comprising n objects;

classifying each object in each second set to generate, for each second set, a corresponding group of attribute values characterizing the objects in the corresponding second set;

selecting the second set of objects corresponding to the group of attribute values best matching the reference group of attribute values; and

identifying the target object in the selected second set of objects in accordance with a position of the target object in the first set of objects.

2. The method of claim 1, wherein the image captured by the 2^nd camera is captured in a time window corresponding to the time of capturing by the 1^st camera the image used for generating the first set of objects.

3. The method of claim 2, wherein the first set of objects are classified using a first set of trained classifiers useable for identifying objects captured by the first camera and each second set of objects are classified using a second set of trained classifiers different from the first set of trained classifiers and useable for identifying objects captured by the second camera.

4. The method of claim 3, wherein the first set of trained classifiers and the second set of trained classifiers are stored in a non-transitory computer readable memory operatively coupled to the processor.

5. The method of claim 4, wherein the objects are vehicles and the sets of objects are platoons of vehicles.

6. The method of claim 5 wherein the vehicles are classified according to one or more attributes selected from the group consisting of:

attributes related to vehicle size, attributes related to vehicle color, attributes related to vehicle type, attributes related to vehicle shape, and attributes related to vehicle aspect ratio.

7. A system for identifying an object in a group of objects appearing in a plurality of cameras comprising:

a non-transitory computer readable memory; and

a processing unit communicatively coupled to the memory, the processing unit comprising a processor configured to;

generate, using an image captured by a 1^st camera and stored in the memory, a first set of n objects comprising the target object and (n−

1) other objects;

classify each object in the first set to generate a reference group of attribute values characterizing the n objects;

generate, using an image captured by a 2^nd camera and stored in the memory, a plurality of second sets of objects, each second set of objects comprising n objects;

classify each object in each second set to generate, for each second set, a corresponding group of attribute values characterizing the objects in the corresponding second set;

select the second set of objects corresponding to the group of attribute values best matching the reference group of attribute values; and

identify the target object in the selected second set of objects in accordance with a position of the target object in the first set of objects.

8. The system of claim 7, wherein the image captured by the 2^nd camera is captured in a time window corresponding to the time of capturing by the 1^st camera the image used for generating the first set of objects.

9. The system of claim 8, wherein the first set of objects are classified using a first set of trained classifiers useable for identifying objects captured by the first camera and each second set of objects are classified using a second set of trained classifiers different from the first set of trained classifiers and useable for identifying objects captured by the second camera.

10. The system of claim 9, wherein the first set of trained classifiers and the second set of trained classifiers are stored in the memory.

11. The system of claim 10, wherein the objects are vehicles and the sets of objects are platoons of vehicles.

12. The system of claim 11, wherein the vehicles are classified according to one or more attributes selected from the group consisting of:

attributes related to vehicle size, attributes related to vehicle color, attributes related to vehicle type, attributes related to vehicle shape, and attributes related to vehicle aspect ratio.

13. A non-transitory storage medium comprising instructions that when executed by a processor, cause the processor to:

generate, using an image captured by a 1^st camera, a first set of n objects comprising the target object and (n−

1) other objects;

classify each object in the first set to generate a reference group of attribute values characterizing the n objects;

generate, using an image captured by a 2^nd camera, a plurality of second sets of objects, each second set of objects comprising n objects;

classify each object in each second set to generate, for each second set, a corresponding group of attribute values characterizing the objects in the corresponding second set;

select the second set of objects corresponding to the group of attribute values best matching the reference group of attribute values; and

identify the target object in the selected second set of objects in accordance with a position of the target object in the first set of objects.

14. The medium of claim 13, wherein the image captured by the 2^nd camera is captured in a time window corresponding to the time of capturing by the 1^st camera the image used for generating the first set of objects.

15. The medium of claim 14, wherein the first set of objects are classified using a first set of trained classifiers useable for identifying objects captured by the first camera and each second set of objects are classified using a second set of trained classifiers different from the first set of trained classifiers and useable for identifying objects captured by the second camera.

16. The medium of claim 15, wherein the first set of trained classifiers and the second set of trained classifiers are stored in a non-transitory computer readable memory operatively coupled to the processor.

17. The medium of claim 16, wherein the objects are vehicles and the sets of objects are platoons of vehicles.

18. The medium of claim 17, wherein the vehicles are classified according to one or more attributes selected from the group consisting of:

attributes related to vehicle size, attributes related to vehicle color, attributes related to vehicle type, attributes related to vehicle shape, and attributes related to vehicle aspect ratio.