Method and system for detection-based segmentation-free license plate recognition

• US 10,068,146 B2
• Filed: 02/25/2016
• Issued: 09/04/2018
• Est. Priority Date: 02/25/2016
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Inventors: Bulan, Orhan, Ramesh, Palghat, Kozitsky, Vladimir
• Primary Examiner(s): Wang, Carol
• Application Number:
  US15/052,941
• Publication Number:
  US 20170249524A1
• Time in Patent Office: 922 Days
• CPC Class Codes: G06F 18/2193 based on specific statistical tests
• CPC Class Codes: G06F 18/295 Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models
• CPC Class Codes: G06V 10/85 Markov-related models; Markov random fields
• 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: G06V 20/62 Text, e.g. of license plates, overlay texts or captions on TV images
• CPC Class Codes: G06V 20/625 License plates
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/1463 Orientation detection or correction, e.g. rotation of multiples of 90 degrees

Abstract

A detection-based segmentation-free method and system for license plate recognition. An image of a vehicle is initially captured utilizing an image-capturing unit. A license plate region is located in the image of the vehicle. A set of characters can then be detected in the license plate region and a geometry correction performed based on a location of the set of characters detected in the license plate region. An operation for sweeping an OCR across the license plate region can be performed to infer characters with respect to the set of characters and locations of the characters utilizing a hidden Markov model and leveraging anchored digit/character locations.

Claims

1. A detection-based segmentation-free method for license plate recognition, said method comprising:

detecting a discrete set of characters in a localized license plate region in an image of a vehicle;

initially locating said license plate region in said image of said vehicle prior to said detecting said discrete set of characters in a localized license plate region in said image of said vehicle;

performing a geometry correction based on a location of said set of characters detected in said localized license plate region;

sweeping an optical character recognition (OCR) across said localized license plate region to infer detected and undetected characters in said localized license plate region in said image with respect to said set of characters and detected locations of said characters;

utilizing a HMM (hidden Markov model) to infer said characters, wherein detected digit locations in a Viterbi decoding facilitates character location in a framework of said HMM, wherein at least one character among said set of characters comprises at least one digit among a plurality of digits and wherein digit locations with respect to said plurality of digits are anchored in said sweeping OCR process and identify spacing between characters and improve decoding performance; and

incorporating detected character types with respect to said characters and said detected locations of said characters as a part of said sweeping said OCR.

2. The method of claim 1 further comprising capturing said image of said vehicle with an image-capturing unit.

3. The method of claim 1 wherein said OCR comprises a segmentation-free OCR utilizing said HMM.

4. The method of claim 3 wherein said performing said geometry correction based on said location of said set of characters detected in said localized license plate region, further comprises:

performing a vertical cropping operation with respect to said set of characters detected in said localized license plate region.

5. A detection-based segmentation-free system for license plate recognition, said system comprising:

at least one processor; and

a computer-usable medium embodying computer program code, said computer-usable medium capable of communicating with said at least one processor, said computer program code comprising instructions executable by said at least one processor and configured for;

capturing an image of a vehicle with an image-capturing unit;

detecting a discrete set of characters in a localized license plate region in said image of said vehicle;

initially locating said license plate region in said image of said vehicle prior to said detecting said discrete set of characters in a localized license plate region in said image of said vehicle;

performing a geometry correction based on a location of said set of characters detected in said localized license plate region;

sweeping an optical character recognition (OCR) across said localized license plate region to infer detected and undetected characters in said localized license plate region in said image with respect to said set of characters and detected locations of said characters;

utilizing a HMM (hidden Markov model) to infer said characters, wherein detected digit locations in a Viterbi decoding facilitates character location in a framework of said HMM, wherein at least one character among said set of characters comprises at least one digit among a plurality of digits and wherein digit locations with respect to said plurality of digits are anchored in said sweeping OCR process and identify spacing between characters and improve decoding performance; and

incorporating detected character types with respect to said characters and said detected locations of said characters as a part of said sweeping said OCR.

6. The system of claim 5 wherein said OCR comprises a segmentation-free OCR utilizing said HMM.

7. The system of claim 6 wherein said instructions for performing said geometry correction based on said location of said set of characters detected in said localized license plate region are further configured for:

performing a vertical cropping operation with respect to said set of characters detected in said localized license plate region.

8. A non-transitory processor-readable medium storing computer code representing instructions to cause a process for detection-based segmentation-free license plate recognition, said computer code comprising code to:

capture an image of a vehicle with an image-capturing unit;

detect a discrete set of characters in a localized license plate region in said image of said vehicle;

initially locate said license plate region in said image of said vehicle prior to said detecting said discrete set of characters in a localized license plate region in said image of said vehicle;

perform a geometry correction based on a location of said set of characters detected in said localized license plate region;

sweep an optical character recognition (OCR) across said localized license plate region to infer detected and undetected characters in said localized license plate region in said image with respect to said set of characters and detected locations of said characters;

utilize a HMM (hidden Markov model) to infer said characters, wherein detected digit locations in a Viterbi decoding facilitates character location in a framework of said HMM, wherein at least one character among said set of characters comprises at least one digit among a plurality of digits and wherein digit locations with respect to said plurality of digits are anchored in said sweeping OCR process and identify spacing between characters and improve decoding performance; and

incorporate detected character types with respect to said characters and said detected locations of said characters as a part of said sweeping said OCR.

9. The processor-readable medium of claim 8 wherein said OCR comprises a segmentation-free OCR utilizing said HMM.