Adaptive character segmentation method and system for automated license plate recognition

• US 9,042,647 B2
• Filed: 06/06/2013
• Issued: 05/26/2015
• Est. Priority Date: 06/06/2013
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Kozitsky, Vladimir, Burry, Aaron, Fillion, Claude
• Primary Examiner(s): COUSO, JOSE L
• Application Number:
  US13/911,448
• Publication Number:
  US 20140363052A1
• Time in Patent Office: 719 Days
• US Class Current: 382/171
• CPC Class Codes: G06V 10/26 Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
• CPC Class Codes: G06V 20/62 Text, e.g. of license plates, overlay texts or captions on TV images
• CPC Class Codes: G06V 20/625 License plates

Abstract

Methods, systems and processor-readable media for adaptive character segmentation in an automatic license plate recognition application. A region of interest can be identified in an image of a license plate acquired via an automatic license plate recognition engine. Characters in the image with respect to the region of interest can be segmented using a histogram projection associated with particular segmentation threshold parameters. The characters in the image can be iteratively validated if a minimum number of valid characters is determined based on the histogram projection and the particular segmentation threshold parameters to produce character images sufficient to identify the license plate.

Claims

1. A method for adaptive character segmentation in an automatic license plate recognition application, said method comprising:

identifying via at least one processor a region of interest in an image of a license plate acquired via an automatic license plate recognition engine that communicates with said at least one processor;

segmenting via said at least one processor said characters in said image with respect to said region of interest using a histogram projection associated with particular segmentation threshold parameters;

iteratively validating said characters in said image via said at least one processor if a minimum number of valid characters is determined based on said histogram projection and said particular segmentation threshold parameters to produce character images sufficient to identify said license plate; and

increasing said particular segmentation threshold parameters in small steps to avoid over-segmentation.

2. The method of claim 1 further comprising increasing said histogram projection by cutting said particular segmentation threshold parameters and repeating said segmenting of said characters and said validating said characters.

3. The method of claim 2 further comprising performing a segmentation error recovery operation with respect to said characters iteratively validated.

4. The method of claim 3 further comprising outputting at least one character image iteratively validated and associated with said image to an optical character recognition engine for further processing.

5. The method of claim 3 further comprising using feedback from an output metric to increase an aggressiveness of said particular segmentation threshold parameters until a result that is sufficiently believable to validate said characters.

6. The method of claim 5 further comprising determining if an overall confidence threshold is met with respect to iteratively validating said characters and assist in identifying said license plate.

7. The method of claim 1 further comprising:

increasing said histogram projection by cutting said particular segmentation threshold parameters and repeating said segmenting of said characters and said validating said characters; and

performing a segmentation error recovery operation with respect to said characters iteratively validated.

8. A system for adaptive character segmentation in an automatic license plate recognition application, said system comprising:

at least one processor;

a data bus coupled to said at least one processor; and

a computer-usable medium embodying computer program code, said computer-usable medium being coupled to said data bus, said computer program code comprising instructions executable by said at least one processor and configured for;

identifying a region of interest in an image of a license plate acquired via an automatic license plate recognition engine that communicates with said at least one processor;

segmenting said characters in said image with respect to said region of interest using a histogram projection associated with particular segmentation threshold parameters;

iteratively validating said characters in said image if a minimum number of valid characters is determined based on said histogram projection and said particular segmentation threshold parameters to produce character images sufficient to identify said license plate; and

increasing said particular segmentation threshold parameters in small steps to avoid over-segmentation.

9. The system of claim 8 wherein said instructions are further configured for increasing said histogram projection by cutting said particular segmentation threshold parameters and repeating said segmenting of said characters and said validating said characters.

10. The system of claim 8 wherein said instructions are further configured for processing a segmentation error recovery operation with respect to said characters iteratively validated.

11. The system of claim 10 wherein said instructions are further configured for outputting at least one character image iteratively validated and associated with said image to an optical character recognition engine for further processing.

12. The system of claim 11 wherein said instructions are further configured for using feedback from an output metric to increase an aggressiveness of said particular segmentation threshold parameters until a result that is sufficiently believable to validate said characters.

13. The system of claim 10 wherein said instructions are further configured for determining if an overall confidence threshold is met with respect to iteratively validating said characters and assist in identifying said license plate.

14. The system of claim 8 wherein said instructions are further configured for:

increasing said histogram projection by cutting said particular segmentation threshold parameters and repeating said segmenting of said characters and said validating said characters; and

processing a segmentation error recovery operation with respect to said characters iteratively validated.

15. A processor-readable medium storing computer code representing instructions to cause a process for adaptive character segmentation in an automatic license plate recognition application, said computer code comprising computer code to:

identify a region of interest in an image of a license plate acquired via an automatic license plate recognition engine;

segment said characters in said image with respect to said region of interest using a histogram projection associated with particular segmentation threshold parameters;

iteratively validate said characters in said image if a minimum number of valid characters is determined based on said histogram projection and said particular segmentation threshold parameters to produce character images sufficient to identify said license plate; and

increase said articular segmentation threshold parameters in small steps to avoid over-segmentation.

16. The processor-readable medium of claim 15 wherein said code further comprises code to increase said histogram projection by cutting said particular segmentation threshold parameters and repeating said segmenting of said characters and said validating said characters.

17. The processor-readable medium of claim 16 wherein said code further comprises code to perform a segmentation error recovery operation with respect to said characters iteratively validated.

18. The processor-readable medium of claim 17 wherein said code further comprises code to output at least one character image iteratively validated and associated with said image to an optical character recognition engine for further processing.

19. The processor-readable medium of claim 17 wherein said code further comprises code to use feedback from an output metric to increase an aggressiveness of said particular segmentation threshold parameters until a result that is sufficiently believable to validate said characters.

20. The processor-readable medium of claim 18 wherein said code further comprises code to determine if an overall confidence threshold is met with respect to iteratively validating said characters and assist in identifying said license plate.