Apparatus Capable of Detecting Location of Object Contained in Image Data and Detection Method Thereof

• US 20130022236A1
• Filed: 09/07/2011
• Published: 01/24/2013
• Est. Priority Date: 07/20/2011
• Status: Active Grant

• Current Assignee: Altek Corporation
• Original Assignee: Altek Corporation
• Inventors: Chang, Yin-Pin, Chang, Wen-Yan, Chou, Hong-Long
• Granted Patent: US 8,538,079 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 7/77 using statistical methods

Abstract

An apparatus capable of detecting location of object contained in image data and its detecting method are disclosed. The apparatus comprises an image capturing module, a weight assignment module, and a processing module. The image capturing module is for capturing an image. The weight assignment module performs the pixel weight/probability assignment according to the priori information and the image, and figures out the initial gravity center of the object according to the object location initialization. The processing module performs the statistical analysis according to the result of the pixel weight/probability assignment and the initial gravity center of the object so as to obtain the analysis result and update the object location. The processing module determines whether or not the analysis result meets the preset value, if it does, the processing module outputs an estimated result; if it doesn'"'"'t, the processing module repeats the foregoing processes.

Claims

1. An apparatus capable of detecting location of object contained in image data comprising:

an image capturing module for capturing image data;

a weight assignment module performing a weight assignment based upon a sub-image corresponding to a possible initial location of an object to be detected within the image data and priori information of the object to be detected, and calculating an initial gravity center of the object to be detected by utilizing the possible initial location; and

a processing module performing a statistical analysis based upon a result of the weight assignment and the initial gravity center to calculate a possible location of the object to be detected and to obtain a gravity center coordinate, and determining whether or not a result of the statistical analysis matches a preset value;

wherein if the result of the statistical analysis matches the preset value, the processing module generates an estimated result contained the possible location of the object to be detected;

if the result of the statistical analysis does not matches the preset value, the processing module replaces the possible initial location with the possible location and calculates the possible location and the gravity center coordinate of the object to be detected once again.

2. The apparatus capable of detecting location of object contained in image data as recited in claim 1, wherein the weight assignment is performed by converting the priori information into a probability density function (PDF) and respectively assigning a weight value for each pixel according to color values, edge orientation values and corresponding values in the PDF of each pixel in the sub-image corresponding to the possible initial location within the image data.

3. The apparatus capable of detecting location of object contained in image data as recited in claim 2, wherein the statistical analysis is performed by utilizing the weight value and calculating a space relationship for each pixel relative to a gravity center so as to compute a weighted covariance matrix of the object to be detected and executing eigen decomposition or singular value decomposition for the weighted covariance matrix.

4. The apparatus capable of detecting location of object contained in image data as recited in claim 1, wherein the preset value is a predetermined precision, a predetermined iterative time or a combination thereof.

5. The apparatus capable of detecting location of object contained in image data as recited in claim 1, wherein the priori information comprises color distribution information, a histogram, an edge orientation and gradient histogram or a combination thereof regarding the object to be detected.

6. The apparatus capable of detecting location of object contained in image data as recited in claim 1, wherein the priori information is generated by training data, manual designation or a clustering manner, and the possible initial location of the object to be detected is generated by the manual designation, the clustering manner or random sample consensus.

7. A method of detecting location of object contained in image data, comprising steps of:

capturing image data;

performing a weight assignment based upon an sub-image corresponding to a possible initial location of an object to be detected within the image data and priori information of the object to be detected, and calculating an initial gravity center of the object to be detected according to the possible initial location;

performing a statistical analysis based upon a result of the weight assignment and the initial gravity center, so as to calculate a possible location of the object to be detected and calculating a gravity center coordinate;

replacing the possible initial location with the possible location through the processing module and repeating foregoing steps until a result of the statistical analysis matches a preset value; and

generating an estimated result contained possible location of the object to be detected.

8. The method of detecting location of object contained in image data as recited in claim 7, wherein the weight assignment is that the priori information is converted into a probability density function (PDF), and a weight value is respectively assigned for each pixel according to color values, edge orientation values and corresponding values in the PDF of each pixel in the sub-image corresponding to the possible initial location within the image data.

9. The method of detecting location of object contained in image data as recited in claim 8, wherein the statistical analysis is performed by utilizing the weight value and calculating a space relationship for each pixel relative to a gravity center so as to compute a weighted covariance matrix of the object to be detected and executing eigen decomposition or singular value decomposition for the weighted covariance matrix.

10. The method of detecting location of object contained in image data as recited in claim 7, wherein the preset value is a predetermined precision, a predetermined iterative time or a combination thereof.

11. The method of detecting location of object contained in image data as recited in claim 7, wherein the priori information comprises color distribution information, a histogram, an edge orientation and gradient histogram or a combination thereof regarding the object to be detected.

12. The method of detecting location of object contained in image data as recited in claim 7, wherein the priori information is generated by training data, manual designation or a clustering manner, and the possible initial location of the object to be detected is generated by the manual designation, the clustering manner or random sample consensus.

13. An apparatus capable of detecting location of object contained in image data comprising:

an image capturing module for capturing image data;

a weight assignment module performing a weight assignment based upon an image within an initial shape limit frame of an object to be detected within the image data and priori information of the object to be detected and for calculating an initial gravity center of the object to be detected by utilizing the initial shape limit frame;

a computation module performing a statistical analysis based upon a result of the weight assignment and the initial gravity center so as to calculate a shape limit frame of the object to be detected and obtain a gravity center coordinate; and

a processing module for determining whether or not a result of the statistical analysis matches a preset value;

wherein if the result of the statistical analysis matches the preset value, the processing module generates an estimated result contained the possible location of the object to be detected;

if the result of the statistical analysis does not matches the preset value, the initial shape limit frame is replaced with the shape limit frame to calculate the shape limit frame and the gravity center coordinate of the object to be detected once again.

14. The apparatus capable of detecting location of object contained in image data as recited in claim 13, wherein the weight assignment is that the priori information is converted into a probability density function (PDF), and a weight value is respectively assigned for each pixel according to color values, edge orientation values and corresponding values in the PDF of each pixel in the image within the initial shape limit frame.

15. The apparatus capable of detecting location of object contained in image data as recited in claim 14, wherein the statistical analysis is performed by utilizing the weight value and calculating a space relationship for each pixel relative to a gravity center so as to compute a weighted covariance matrix of the object to be detected and executing eigen decomposition or singular value decomposition for the weighted covariance matrix.

16. The apparatus capable of detecting location of object contained in image data as recited in claim 13, wherein the preset value is a predetermined precision, a predetermined iterative time or a combination thereof.