Method and system for image identification and identification result output

• US 8,391,560 B2
• Filed: 07/30/2009
• Issued: 03/05/2013
• Est. Priority Date: 04/30/2009
• Status: Active Grant

• Current Assignee: Industrial Technology Research Institute
• Original Assignee: Industrial Technology Research Institute
• Inventors: Lin, Yu-Ting, Chang, Chun-Lung, Wu, Tung-Chuan, Tsai, Ya-Hui, Huang, Kuo-Tang
• Primary Examiner(s): Newman, Michael A
• Application Number:
  US12/512,670
• Publication Number:
  US 20100278436A1
• Time in Patent Office: 1,314 Days
• Field of Search: 382/105, 382/104, 382/176, 382/177, 382/181, 382/190, 382/206, 382/216, 382/218, 382/220, 382/289, 382/290
• US Class Current: 382/105
• CPC Class Codes: G06V 20/625 License plates
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names

Abstract

The present invention provides a method and a system for image identification and identification result output, which determines a location coordinate with respect to an image and a rotating angle based on at least one direction of the image according to features of the image. The image is compared to a plurality of sample images stored in a database according to the rotating angle so as to obtain at least one identification result. By means of the method and the system of the present invention, identification can be achieved with respect to various rotating angles and distances so as to improve the identification rate.

Claims

1. A method for image identification and identification result output, comprising steps of:

providing an image to be identified comprising a region of interest thereon;

acquiring a feature in the region of interest;

determining a rotating angle based on at least one direction according to the feature; and

comparing the image with sample images in a database corresponding to the rotating angle to obtain at least one identification result,

wherein the step of comparing comprises the steps of;

acquiring a feature image from the region of interest;

acquiring the sample images in the database according to the at least one rotating angle, each sample image having respectively a standard image region and at least one non-standard image region, wherein the standard image region has pixels corresponding to a first feature value respectively, and the non-standard image region has pixels corresponding to a second feature value respectively;

performing a normalized correlation matching calculation on a third feature value of each pixel in the feature image and the first feature value or the second feature value corresponding to each pixel in the plurality of sample images to obtain a similarity index of the feature image corresponding to the plurality of sample images respectively;

collecting a plurality of similarity indexes with respect to the feature image compared with the plurality of sample images; and

sorting the plurality of similarity indexes and outputting at least one of comparison results;

wherein the step of acquiring the sample images comprise the steps of;

determining the size of the sample images;

providing in each of the sample images the standard image region;

providing in each of the sample images the non-standard image region;

storing the sample images in the database;

observing the identification result after a plurality times of training; and

modifying the standard image region and the non-standard image region in the sample images; and

wherein the non-standard image region is formed by providing a pixel with the first feature value and the second feature value in the standard image region according to mis-identified characters in the standard image region.

15. The method for image identification and identification result output as recited in claim 1, wherein the step of determining the rotating angle comprises steps of:

determining a standard image comprising a standard feature region;

performing a calculation on the feature in the region of interest and the standard feature region so as to obtain a location corresponding to the standard feature region in the region of interest; and

determining the rotating angle of the region of interest according to the location.

16. The method for image identification and identification result output as recited in claim 15, wherein the calculation is based on geometric matching.

17. The method for image identification and identification result output as recited in claim 1, wherein the rotating angle is based on a first direction and is within a range between ±

60 degrees.

18. The method for image identification and identification result output as recited in claim 1, wherein the rotating angle is based on a second direction and is within a range between ±

60 degrees.

19. The method for image identification and identification result output as recited in claim 1, further comprising the step of determining a scale relation between the region of interest and the sample images according to the feature.

20. The method for image identification and identification result output as recited in claim 1, wherein the region of interest comprises identification information.

21. The method for image identification and identification result output as recited in claim 1, wherein the feature further comprises contrast, greyscale, chrominance or spectrum on the boundary of identification information.

22. The method for image identification and identification result output as recited in claim 1, wherein the feature comprises contrast, greyscale, chrominance or spectrum on the boundary of the region of interest.

23. The method for image identification and identification result output as recited in claim 1, wherein the region of interest is formed by the boundary of a license plate on a carrier.

24. The method for image identification and identification result output as recited in claim 1, wherein the first feature value and the second feature value are respectively a combination of a weight value and a greyscale value, and the third feature value is a greyscale value.

25. The method for image identification and identification result output as recited in claim 1, wherein normalized correlation matching is performed by dividing the product of a weight value corresponding to each pixel in the sample image, the difference between a greyscale value of each pixel and the average greyscale value in the sample image, and the difference between the greyscale value of each pixel and the average greyscale value in the feature image by the product of the standard deviation of the greyscale value of the sample image and the standard deviation of the greyscale value of the feature image.

2. A system for image identification and identification result output, comprising:

a database capable of providing a plurality of sample images, each sample image having respectively a standard image region and at least one non-standard image region, the standard image region having pixels corresponding to a first feature value respectively, and the non-standard image region having pixels corresponding to a second feature value respectively, the non-standard image region being formed by providing a pixel with the first feature value and the second feature value in the standard image region according to mis-identified characters in the standard image region;

an image acquiring unit capable of acquiring an image having a region of interest;

a feature acquiring unit capable of acquiring a feature in the region of interest on the image;

an operation and processing unit capable of determining a rotating angle based on at least one direction in the region of interest according to the feature, acquiring a feature image from the region of interest, acquiring the sample images in the database according to the at least one rotating angle, performing a normalized correlation matching calculation on a third feature value of each pixel in the feature image and the first feature value or the second feature value corresponding to each pixel in the plurality of sample images to obtain a similarity index of the feature image corresponding to the plurality of sample images respectively, collecting a plurality of similarity indexes with respect to the feature image compared with the plurality of sample images, and sorting the plurality of similarity indexes and outputting at least one of comparison results, wherein the operation and processing unit determines the sizes of the sample images, identifies a standard image region and a non-standard image region in each of the sample images, stores the sample images in the database, obtains an identification result after training a plurality of times, and modifies the standard image region and the non-standard image region in the sample images; and

an identification and output unit that is electrically connected to the operation and processing unit to output at least one identification result.

3. The system for image identification and identification result output as recited in claim 2, wherein the operation and processing unit further comprises an enhancing unit capable of enhancing the image.

4. The system for image identification and identification result output as recited in claim 2, wherein the angle is determined by geometric matching.

5. The system for image identification and identification result output as recited in claim 2, wherein the rotating angle based on a first direction is within a range between ±

60 degrees.

6. The system for image identification and identification result output as recited in claim 2, wherein the rotating angle based on a second direction is within a range between ±

60 degrees.

7. The system for image identification and identification result output as recited in claim 2, wherein each sample image has respectively a standard image region and at least one non-standard image region, wherein the standard image region has pixels corresponding to a first feature value, and wherein the non-standard image region has pixels corresponding to a second feature value.

8. The system for image identification and identification result output as recited in claim 7, wherein the non-standard image region is formed by providing a pixel with the first feature value and the second feature value in the standard image region according to mis-identified characters in the standard image region.

9. The system for image identification and identification result output as recited in claim 2, wherein the operation and processing unit excludes images of impossible characters or digits according to a relative location of the feature image in the identification mark when each feature image is compared with the plurality of sample images stored in the database by the operation and processing unit.

10. The system for image identification and identification result output as recited in claim 2, wherein a normalized correlation matching calculation is performed on a third feature value of each pixel in the feature image and the first feature value or the second feature value corresponding to each pixel in the plurality of sample images to obtain a similarity index of the feature image corresponding to the plurality of sample images respectively.

11. The system for image identification and identification result output as recited in claim 2, wherein the region of interest comprises identification information.

12. The system for image identification and identification result output as recited in claim 11, wherein the feature comprises contrast, greyscale, chrominance or spectrum on the boundary of identification information.

13. The system for image identification and identification result output as recited in claim 2, wherein the feature comprises contrast, greyscale, chrominance or spectrum on the boundary of the region of interest.

14. The system for image identification and identification result output as recited in claim 2, wherein the operation and processing unit determines the scale relation between the size of the region of interest and the sample image according to the feature.