License plate recognition

• US 9,405,988 B2
• Filed: 08/06/2014
• Issued: 08/02/2016
• Est. Priority Date: 08/13/2013
• Status: Active Grant

• Current Assignee: Q-Free Netherlands BV (Q-Free ASA)
• Original Assignee: James Alves
• Inventors: Alves, James
• Primary Examiner(s): Patel, Nirav G
• Application Number:
  US14/452,925
• Publication Number:
  US 20150049914A1
• Time in Patent Office: 727 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06V 10/987 with the intervention of an operator
• CPC Class Codes: G06V 20/625 License plates
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names

Abstract

A license plate recognition and image review system and processes are described. The system includes grouping of images that are determined to be of the same vehicle, using an image encoded database such that verification of a license plate read is done through comparison of images of the actual vehicle to images from the encoded database and testing of the accuracy of a manual review process by interspersing previously identified images with real images being reviewed in a batch process.

Claims

1. A method for identifying characters on license plates of a vehicle on a roadway said method comprising:

a) acquiring from an electronic camera, a plurality of electronic images of the vehicle said images including the license plate area,

b) recognizing image features in the electronic images of the vehicle that do not include recognition of the characters of the vehicle'"'"'s license plate,

c) grouping the electronic images that are determined to be of the same vehicle, the determination based upon the image features in the electronic images of the vehicle that do not include recognition of the characters on the vehicle'"'"'s license plate,

d) analyzing a single image from the grouped image to recognize the characters of the license plate,

e) assigning the recognized characters from the single image to all images of the grouped images,

f) calculating a confidence level for the recognition of the characters and if the confidence interval exceeds a preselected value using the recognized characters to identify the vehicle and an owner and if the confidence level is less than a preselected value initiating a manual review of the image for character recognition, and,

g) assigning the confidence level for one image within the group of images to all other images within the group,

h) wherein the steps a)-g) are performed automatically by the computer processor.

2. The method of claim 1 further including a verification process said verification process including determining, by the computer processor, if the recognized characters and identified vehicle have been previously identified on the same roadway and if so automatically verifying the recognized characters and vehicle.

3. The method of claim 1 wherein the manual review includes manually reviewing by a human operator groups of images from different vehicles in a batch of images, said batch of images including known previously identified images interspersed with the images acquired from passing vehicles and calculating the confidence level for the manual review of images acquired from passing vehicles on the basis of the results of review of the known previously identified images wherein the batches of images and the previously identified images are automatically selected by the computer processor.

4. The method of claim 2 wherein the verification process includes comparing, by the computer processor, the at least one acquired image with images in an image database, said image database including at least one of images of vehicle brand emblems, images of vehicle colors, images of vehicle types and images of vehicle models, and said database images being an encoding of vehicle registration information.

5. A computer processor implemented method for automatically identifying characters on license plates of a vehicle on a roadway said method comprising:

a) acquiring from an electronic camera, a plurality of electronic images of the vehicle said images including the license plate area,

b) recognizing image features in the electronic images of the vehicle that do not include recognition of the characters of the vehicle'"'"'s license plate,

c) grouping the electronic images that are determined to be of the same vehicle, the determination based upon the image features of the vehicle that do not include recognition of the characters on the vehicle'"'"'s license plate,

d) analyzing a single image from the grouped image to recognize the characters of the license plate,

e) assigning the recognized characters from the single image to all images of the grouped images,

f) calculating a confidence level for the recognition of the characters and if the confidence interval exceeds a preselected value using the recognized characters to identify the vehicle and an owner and if the confidence level is less than a preselected value initiating a manual review of the image for character recognition,

g) wherein the manual review includes reviewing groups of images from different vehicles in a batch of images, said batch of images including known previously identified images interspersed with the images acquired from passing vehicles and calculating the confidence level for the manual review of images acquired from passing vehicles on the basis of the results of review of the known previously identified images.

6. The method of claim 5 wherein the confidence level for one image within the group of images is assigned by the computer processor to all other images within the group.

7. The method of claim 5 further including a verification process said verification process including determining, by the computer processor, if the recognized characters and identified vehicle have been previously identified on the same roadway and if so automatically verifying the recognized characters and vehicle.

8. The method of claim 7 wherein the verification process includes comparing, by the computer processor, the at least one acquired image with images in an image database, said image database including at least one of images of vehicle brand emblems, images of vehicle colors, images of vehicle types and images of vehicle models, and said database images being an encoding of vehicle registration information.

9. A computer processor implemented method for automatically identifying characters on license plates of a vehicle on a roadway said method comprising:

a) acquiring from an electronic camera a plurality of electronic images of the vehicle said image including the license plate area at multiple points along the roadway, the plurality of images from a single point along the roadway being a transaction,

b) recognizing image features in the electronic images of the vehicle that do not include recognition of the characters of the vehicle'"'"'s license plate, the image features being vehicle signature information,

c) grouping transactions from the multiple points identified as being of the same vehicle identification based upon the image features that do not include character recognition of the characters on the license plate of the vehicle,

d) analyzing the images in each transaction to recognize the characters of the license plate and assigning a confidence level to the recognition of the characters for each of the transactions,

e) verifying the recognition of the characters, verification including comparing the vehicle signature information from the transaction images with vehicle information in a registration database and if the vehicle signature information from the transaction matches the vehicle information in the vehicle registration database verifying the correctness of the character recognition for the grouped transactions,

f) calculating a confidence level for the recognition of the characters and if the confidence interval exceeds a preselected value using the recognized characters to identify the vehicle and an owner and if the confidence level is less than a preselected value initiating a manual review of the image for character recognition, and,

g) assigning the confidence level for one transaction within the group of transactions to all other transactions within the group.

10. The method of claim 9 wherein the manual review includes manually reviewing by a human operator groups of transactions from different vehicles in a batch of transactions, said batch of transactions including known previously identified transactions interspersed with the transactions acquired from passing vehicles and calculating the confidence level for the manual review of transactions acquired from passing vehicles on the basis of the results of review of the known previously identified transactions wherein the batches of transactions and the previously identified transactions are automatically selected by the computer processor.

11. A computer processor implemented method for automatically identifying characters on license plates of a vehicle on a roadway said method comprising:

a) acquiring from an electronic camera, a plurality of electronic images of the vehicle said image including the license plate area at multiple points along the roadway, the plurality of images from a single point along the roadway being a transaction,

b) recognizing image features in the electronic images of the vehicle that do not include recognition of the characters of the vehicle'"'"'s license plate, the image features being vehicle signature information,

c) grouping transactions from the multiple points identified as being of the same vehicle identification based upon image features that does not include character recognition of the characters on the license plate of the vehicle,

d) analyzing the images in each transaction to recognize the characters of the license plate and assigning a confidence level to the recognition of the characters for each of the transactions,

e) verifying the recognition of the characters, verification including comparing the vehicle signature information from the transaction images with vehicle information in a registration database and if the vehicle signature information from the transaction matches the vehicle information in the vehicle registration database verifying the correctness of the character recognition for the grouped transactions,

f) calculating a confidence level for the recognition of the characters and if the confidence interval exceeds a preselected value using the recognized characters to identify the vehicle and an owner and if the confidence level is less than a preselected value initiating a manual review of the image for character recognition, and,

g) wherein the manual review includes reviewing groups of transactions from different vehicles in a batch of transactions, said batch of transactions including known previously identified transactions interspersed with the transactions acquired from passing vehicles and calculating the confidence level for the manual review of transactions acquired from passing vehicles on the basis of the results of review of the known previously identified transactions.

12. The method of claim 11 wherein the confidence level for one transaction within the group of transactions is assigned by the computer processor to all other transactions within the group.