Data augmentation method and system for improved automatic license plate recognition

• US 9,224,058 B2
• Filed: 04/05/2013
• Issued: 12/29/2015
• Est. Priority Date: 04/05/2013
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Paul, Peter, Kozitsky, Vladimir, Burry, Aaron Michael, Hoover, Martin Edward, Cantelli, Mark
• Primary Examiner(s): Bella, Matthew
• Assistant Examiner(s): SHIN, SOO JUNG
• Application Number:
  US13/857,657
• Publication Number:
  US 20140301606A1
• Time in Patent Office: 998 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06V 10/141 Control of illumination
• CPC Class Codes: G06V 20/625 License plates
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names

Abstract

Methods, systems, and processor-readable media for data augmentation utilized in an automatic license plate recognition engine. A machine-readable code can be associated with an automatic license plate recognition engine. The machine-readable code can be configured to define parameters that drive processing within the automatic license plate recognition engine to produce recognition results thereof and enhance a machine readability of a license plate recognized and analyzed via the automatic license plate recognition engine.

Claims

1. A method for data augmentation utilized in an automatic license plate recognition engine, said method comprising:

associating a machine-readable code with an automatic license plate recognition engine that reads and recognizes an image of a license plate of a vehicle captured via at least one camera and which communicates with a barcode reading engine, an image capture engine and a processor that communicates electronically with said at least one camera, at least one illuminator and at least one trigger wherein said at least one trigger comprises a ground loop trigger;

controlling said at least one trigger, said at least one illuminator and said at least one camera via said image capture engine associated with said automatic license plate recognition engine so that said at least one trigger controls said at least one illuminator to illuminate a license plate region of said vehicle;

configuring said machine-readable code to comprise 24 or fewer bits, which define parameters that drive processing within said automatic license plate recognition engine with respect to said processor to produce recognition results thereof and enhance a machine readability of said image of said license plate captured by said at least one camera and recognized and analyzed via said automatic license plate recognition engine; and

locating a barcode region in said image of said license plate via said barcode reading engine using a priori knowledge of said location and edge references and known pixel-to-edge distance relationships to determine a parameter set from said parameters, which is sent to said license plate reading engine, which uses said parameter set to determine a correct character sequence and a state jurisdiction to determine a unique identifier for said vehicle; and

periodically updating said parameters based on a current operation of said automatic license plate recognition engine and historical statistics associated with said automatic license plate recognition engine.

2. The method of claim 1 further comprising:

decoding said machine-readable code to determine said parameter set including said parameters;

encoding said machine-readable code with information based on a hierarchy having hierarchy levels with higher levels that are more robust to image noises and lower levels that are less robust with respect to said image noises and which contain more details at said lower levels; and

arranging said hierarchy with augmented data at a highest level in said hierarchy such that said higher levels and said lower levels further define said parameters that drive said processing within said automatic license plate recognition engine to produce said recognition results thereof and enhance a machine readability of said license plate recognized and analyzed via said automatic license plate recognition engine.

3. The method of claim 2 wherein said machine readable code is encoded with information based on known failure modes of said automatic license plate recognition engine.

4. The method of claim 3 wherein said parameters include data indicative of whether or not said license plate contains a potential for misreading due to confusing characters.

5. The method of claim 3 wherein said parameters include data indicative of whether or not said license plate comprises a standard plate or a non-standard plate.

6. The method of claim 5 wherein said parameters include data indicative of whether said license plate contains characters subject to plate frame occlusion.

7. The method of claim 6 wherein said parameters include data indicative of said state jurisdiction.

8. The method of claim 7 wherein said parameters include data indicative of whether or not said license plate possesses a logo.

9. The method of claim 8 wherein said parameters include data indicative of whether the license plate has stacked characters or not.

10. The method of claim 8 wherein said parameters include data indicative of a particular font utilized by said license plate.

11. The method of claim 8 wherein said parameters include data indicative of optimal RGB-to-Gray mapping.

12. The method of claim 11 wherein said parameters include data indicative of the number of characters on said license plate.

13. The method of claim 12 wherein said parameters include data indicative of whether each character on said license plate is a letter or a number.

14. A system for data augmentation utilized in an automatic license plate recognition engine, said system comprising:

at least one camera that communicates electronically with at least one illuminator and at least one trigger;

an automatic license plate recognition engine that reads and recognizes an image of a license plate of a vehicle captured via said at least one camera and which communicates with a barcode reading engine, an image capture engine;

at least one processor that communicates electronically with said at least one camera, said at least one illuminator and said at least one trigger;

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;

associating a machine-readable code with said automatic license plate recognition engine to read and recognize said image of said license plate of said vehicle captured via said at least one camera and which communicates with said barcode reading engine, said image capture engine and said at least one processor to communicate electronically with said at least one camera, said at least one illuminator and said at least one trigger;

controlling said at least one trigger, said at least one illuminator and said at least one camera via said image capture engine so that said at least one trigger controls said at least one illuminator to illuminate a license plate region of said vehicle;

configuring said machine-readable code to comprise 24 or fewer bits, which define parameters that drive processing within said automatic license plate recognition engine with respect to said at least one processor to produce recognition results thereof and enhance a machine readability of said image of said license plate captured by said at least one camera and recognized and analyzed via said automatic license plate recognition engine;

locating a barcode region in said image of said license plate via said barcode reading engine using a priori knowledge of said location and edge references and known pixel-to-edge distance relationships to determine a parameter set from said parameters, which is sent to said license plate reading engine, which uses said parameter set to determine a correct character sequence and a state jurisdiction to determine a unique identifier for said vehicle; and

periodically updating said parameters based on a current operation of said automatic license plate recognition engine and historical statistics associated with said automatic license plate recognition engine.

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

decoding said machine-readable code to determine said parameter set including said parameters;

encoding said machine-readable code with information based on a hierarchy having hierarchy levels with higher levels that are more robust to image noises and lower levels that are less robust with respect to said image noises and which contain more details at said lower levels; and

arranging said hierarchy with augmented data at a highest level in said hierarchy such that said higher levels and said lower levels further define said parameters that drive said processing within said automatic license plate recognition engine to produce said recognition results thereof and enhance a machine readability of said license plate recognized and analyzed via said automatic license plate recognition engine.

16. The system of claim 15 wherein said machine readable code is encoded with information based on known failure modes of said automatic license plate recognition engine.

17. The system of claim 16 wherein said parameters include data indicative of:

whether or not said license plate contains a potential for misreading due to confusing characters;

whether or not said license plate comprises a standard plate or a non-standard plate; and

whether said license plate contains characters subject to plate frame occlusion.

18. The system of claim 17 wherein said parameters further include data indicative of:

said state jurisdiction;

optimal RGB-to-Gray mapping;

the number of characters on said license plate;

whether each character on said license plate is a letter or a number;

whether or not said license plate possesses a logo;

whether said license plate includes stacked characters; and

a particular font utilized by said license plate.

19. A system for data augmentation utilized in an automatic license plate recognition engine, said system comprising

at least one camera that communicates electronically with at least one illuminator and at least one trigger;

an automatic license plate recognition engine that reads and recognizes an image of a license plate of a vehicle captured via said at least one camera and which communicates with a barcode reading engine and an image capture engine;

at least one processor that communicates electronically with said at least one camera, said at least one illuminator and said at least one trigger, wherein said automatic license plate recognition engine reads and recognizes said image of said license plate of said vehicle captured via said at least one camera and which communicates with said barcode reading engine, said image capture engine and said at least one processor to communicate electronically with said at least one camera, said at least one illuminator and said at least one least one trigger, wherein said at least one trigger controls said at least one illuminator to illuminate a license plate region of said vehicle;

wherein parameters comprising 24 or fewer bits are defined, which drive processing with respect to said at least one processor to produce recognition results thereof and enhance a machine readability of said image of said license plate captured by said at least one camera and recognized and analyzed via said automatic license plate recognition engine;

wherein a barcode region in said image of said license plate is located via said barcode reading engine using a priori knowledge of said location and edge references and known pixel-to-edge distance relationships to determine a parameter set from said parameters, which is sent to said license plate reading engine, which uses said parameter set to determine a correct character sequence and a state jurisdiction to determine a unique identifier for said vehicle; and

wherein said parameters are periodically updated based on a current operation of said automatic license plate recognition engine and historical statistics associated with said automatic license plate recognition engine.

20. The system of claim 19 wherein said at least one processor uses augmented data in a hierarchy having hierarchy levels with higher levels that are more robust to image noises and lower levels that are less robust with respect to said image noises and which contain more details at said lower levels and wherein said hierarchy is arranged with said augmented data at a highest level in said hierarchy such that said higher levels and said lower levels further define said parameters that drive said processing within said automatic license plate recognition engine to produce said recognition results thereof and enhance a machine readability of said license plate recognized and analyzed via said automatic license plate recognition engine and wherein said augmented data comprises an augmentation data bit format that encodes extra information used to improve the performance of said automatic license plate recognition engine.