Object matching method and camera system with an object matching function

• US 10,186,042 B2
• Filed: 11/14/2016
• Issued: 01/22/2019
• Est. Priority Date: 11/20/2015
• Status: Active Grant

• Current Assignee: Vivotek, Inc.
• Original Assignee: Vivotek, Inc.
• Inventors: Liu, Cheng-Chieh, Liu, Wei-Ming
• Primary Examiner(s): Entezari, Michelle M
• Application Number:
  US15/350,123
• Publication Number:
  US 20170148175A1
• Time in Patent Office: 799 Days
• Field of Search: None
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/292 Multi-camera tracking
• 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 23/90 Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums

Abstract

An object matching method is applied to a camera system with an object matching function. The object matching method includes detecting a moving object within overlapped monitoring areas of two image capturing units to generate a first coordinate point and a second coordinate point respectively upon a first coordinate system and a second coordinate system, calculating at least one transforming parameter of the first coordinate system relative to the second coordinate system, acquiring a reliability level according to a comparison result between the transforming parameter and a threshold, determining a final transform parameter by the reliability level, utilizing the final transform parameter to transform the first coordinate point into a third coordinate point upon the second coordinate system, and determining whether the first coordinate point and the second coordinate point indicate the same object by difference between the third coordinate point and the second coordinate point.

Claims

1. An object matching method of matching an object staying inside an overlapped monitoring region between adjacent monitoring regions captured by two image capturing units, the two image capturing units respectively having a first coordinate system and a second coordinate system, the object matching method being applied to a camera system having an operation processor electrically connected with the two image capturing units, the object matching method comprising:

the operation processor detecting at least one moving object located inside the overlapped monitoring region to generate a first coordinate point upon the first coordinate system and a second coordinate point upon the second coordinate system;

the operation processor computing a plurality of transforming parameters of the first coordinate system relative to the second coordinate system;

the operation processor setting some of the plurality of transforming parameters having similar parameter values as being selfsame according to a comparison result between the transforming parameters and a threshold;

the operation processor defining an amount of the said selfsame transforming parameters as a reliability level;

the operation processor determining a final transform parameter by the reliability level;

the operation processor utilizing the final transform parameter to transform the first coordinate point into a third coordinate point upon the second coordinate system; and

the operation processor determining whether the first coordinate point and the second coordinate point are considered as the same moving object when a difference between the third coordinate point and the second coordinate point is smaller than a specific value.

2. The object matching method of claim 1, wherein while a plurality of moving objects is located inside the overlapped monitoring region, any first coordinate point upon the first coordinate system is corresponding to any related second coordinate point upon the second coordinate system to make a pair, and a difference between coordinate points of each pair or a combination of the said difference is utilized to determine whether the two coordinate points respectively upon different coordinate systems are considered as the same moving object.

3. The object matching method of claim 2, wherein while the two coordinate points respectively upon different coordinate systems are considered as the same moving object, the moving object is marked by a new reference number.

4. The object matching method of claim 3, wherein routes of the two coordinate points respectively upon different coordinate systems about the same moving object are combined to generate a moving route of the moving object marked by the new reference number.

5. The object matching method of claim 1, wherein the image capturing units are driven to simultaneously capture monitoring images of the moving object within the overlapped monitoring region.

6. The object matching method of claim 1, wherein a step of computing the plurality of transforming parameters of the first coordinate system relative to the second coordinate system comprises:

the operation processor acquiring coordinate values of the moving object at several time points during at least one predetermined time interval;

the operation processor computing an assuming parameter by the coordinate values;

the operation processor generating an assuming function according to the assuming parameter;

the operation processor substituting the coordinate values of the moving object at the whole time points during the at least one predetermined time interval for the assuming function to generate at least one substitution result; and

the operation processor determining whether the assuming parameter corresponding to the substitution result is regarded as the transforming parameters in accordance with the substitution result.

7. The object matching method of claim 6, wherein a period of the moving object staying inside the overlapped monitoring region is divided into a plurality of predetermined time intervals.

8. The object matching method of claim 6, wherein a step of determining whether the assuming parameter corresponding to the substitution result is regarded as the transforming parameters in accordance with the substitution result is:

defining the assuming parameter having a maximal amount of the substitution results within an allowed range as the transforming parameters.

9. The object matching method of claim 1, wherein when the reliability level is greater than a predetermined value, some of the plurality of transforming parameters having the similar parameter values are utilized to define the final transform parameter.

10. The object matching method of claim 1, wherein the transforming parameters is selected from a group consisting of an offset, a scaling ratio, a rotary view angle between the first coordinate system and the second coordinate system, and combination thereof.

11. A camera system with an object matching function, having an operation processor and two image capturing units electrically connected with each other, an overlapped monitoring region being set between two adjacent monitoring regions captured by the two image capturing units, the two image capturing units respectively having a first coordinate system and a second coordinate system, the operation processor being adapted to detect at least one moving object located inside the overlapped monitoring region to generate a first coordinate point upon the first coordinate system and a second coordinate point upon the second coordinate system, to computing a plurality of transforming parameters of the first coordinate system relative to the second coordinate system, to set some of the plurality of transforming parameters having similar parameter values as being selfsame according to a comparison result between the transforming parameters and a threshold, to define an amount of the said selfsame transforming parameters as a reliability level, to determine a final transform parameter by the reliability level, to utilize the final transform parameter to transform the first coordinate point into a third coordinate point upon the second coordinate system, and to determine whether the first coordinate point and the second coordinate point are considered as the same moving object when a difference between the third coordinate point and the second coordinate point is smaller than a specific value, for matching the same moving object staying inside the overlapped monitoring region.

12. The camera system of claim 11, wherein while a plurality of moving objects is located inside the overlapped monitoring region, any first coordinate point upon the first coordinate system is corresponding to any related second coordinate point upon the second coordinate system to make a pair, and a difference between coordinate points of each pair or a combination of the said difference is utilized to determine whether the two coordinate points respectively upon different coordinate systems are considered as the same moving object.

13. The camera system of claim 12, wherein while the two coordinate points respectively upon different coordinate systems are considered as the same moving object, the moving object is marked by a new reference number.

14. The camera system of claim 13, wherein routes of the two coordinate points respectively upon different coordinate systems about the same moving object are combined to generate a moving route of the moving object marked by the new reference number.

15. The camera system of claim 11, wherein the operation processor is further adapted to acquire coordinate values of the moving object at several time points during at least one predetermined time interval, to compute an assuming parameter by the coordinate values, to generate an assuming function according to the assuming parameter, to substitute the coordinate values of the moving object at the whole time points during the at least one predetermined time interval for the assuming function to generate at least one substitution result, and to determine whether the assuming parameter corresponding to the substitution result is regarded as the transforming parameters in accordance with the substitution result while computing the transforming parameters of the first coordinate system relative to the second coordinate system.

16. The camera system of claim 15, wherein the operation processor is further adapted to define the assuming parameter having a maximal amount of the substitution results within an allowed range as the transforming parameters while determining whether the assuming parameter corresponding to the substitution result is regarded as the transforming parameters in accordance with the substitution result.

17. The camera system of claim 11, wherein when the reliability level is greater than a predetermined value, some of the plurality of transforming parameters having the similar parameter values are utilized to define the final transform parameter.

18. The camera system of claim 11, wherein the transforming parameters is selected from a group consisting of an offset, a scaling ratio, a rotary view angle between the first coordinate system and the second coordinate system, and combination thereof.