Delayed vehicle identification for privacy enforcement

• US 9,412,031 B2
• Filed: 10/16/2013
• Issued: 08/09/2016
• Est. Priority Date: 10/16/2013
• Status: Active Grant

• Current Assignee: Conduent Business Services, LLC (Conduent, Inc.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Ciriza, Victor, Rodriguez-Serrano, Jose Antonio, Perronnin, Florent C., Poirier, Herve, Roulland, Frederic
• Primary Examiner(s): Park, Chan
• Assistant Examiner(s): Thomas, Mia M
• Application Number:
  US14/054,998
• Publication Number:
  US 20150104073A1
• Time in Patent Office: 1,028 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/5846 using extracted text
• CPC Class Codes: G06F 18/2415 based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
• CPC Class Codes: G06V 10/758 Involving statistics of pixels or of feature values, e.g. histogram matching
• CPC Class Codes: G06V 10/764 using classification, e.g. of video objects
• CPC Class Codes: G06V 20/56 exterior to a vehicle by using sensors mounted on the vehicle
• CPC Class Codes: G06V 20/625 License plates
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names
• CPC Class Codes: G06V 2201/10 Recognition assisted with metadata

Abstract

A method for recognition of an identifier such as a license plate includes storing first visual signatures, each extracted from a first image of a respective object, such as a vehicle, captured at a first location, and first information associated with the first captured image, such as a time stamp. A second visual signature is extracted from a second image of a second object captured at a second location and second information associated with the second captured image is acquired. A measure of similarity is computed between the second visual signature and at least some of the first visual signatures to identify a matching one. A test is performed, which is a function of the first and the second information associated with the matching signatures. Only when it is confirmed that the test has been met, identifier recognition is performed to identify the identifier of the second object.

Claims

1. A method for recognition of an identifier, comprising:

for each of a plurality of objects, storing a first visual signature, which has been extracted from at least a part of a respective first image of the object captured at a first location, the first visual signature comprising a multidimensional statistical representation of pixels of at least the part of the respective first image, and first information associated with the first captured image, the first information comprising a first time stamp;

computing a measure of similarity between a second visual signature and at least some of the first visual signatures to identify a matching one of the first visual signatures, based on the computed measure of similarity, the second visual signature having been extracted from at least a part of a second image of a second object captured at a second location, based on at least a part of the second image, the second visual signature comprising a multidimensional statistical representation of pixels of at least the part of the second image;

retrieving the stored first information associated with the matching one of the first visual signatures and stored second information associated with the second captured image, the second information comprises a second time stamp which is later in time than the first time stamp;

performing a test which is a function of the retrieved stored first time stamp and the stored second time stamp; and

recognizing an identifier of the second object based on a captured image of the second object, only when it is confirmed that the test has been met.

2. The method of claim 1, wherein at least one of the computing the measure of similarity, performing the test, and the recognizing of the identifier is performed with a computer processor.

3. The method of claim 1, further comprising:

for each of the plurality of objects;

receiving the first image of the object captured at the first location and the information associated with the first captured image,

extracting the first visual signature based on at least a part of the first image, and

storing the first visual signature and the information associated with the first image.

4. The method of claim 3, further comprising deleting or encrypting each first image after extracting the respective visual signature and at least prior to the recognizing of the identifier.

5. The method of claim 1, further comprising:

receiving the second image of the second object captured at the second location and the information associated with the second image;

extracting the second visual signature based on at least a part of the second image.

6. The method of claim 1, wherein each object comprises a vehicle and each identifier comprises a sequence of characters.

7. The method of claim 6, wherein the recognizing of the identifier comprises performing automated license plate recognition on the first or second image.

8. The method of claim 1, wherein the performing of the test comprises computing a speed based on the first and second time stamps and a distance between the first and second locations and comparing the computed speed to a threshold speed.

9. The method of claim 8, wherein the test is met when the computed speed exceeds the threshold speed.

10. The method of claim 1, wherein the performing of the test comprises comparing a difference between the first and second time stamps with an authorized time.

11. The method of claim 10, wherein the test is met when the difference between the first and second time stamps exceeds the authorized time by a predetermined amount.

12. The method of claim 1, further comprising implementing a process based on the recognized identifier.

13. The method of claim 1, wherein first visual signatures are stored in a database and the method further comprises purging first visual signatures from the database when the test can no longer be met.

14. The method of claim 1, wherein the second object is one of the plurality of objects.

15. The method of claim 1, wherein when the test has not been met no recognition of the identifier is performed.

16. A computer program product comprising a non-transitory recording medium storing instructions, which when executed on a computer causes the computer to perform the method of claim 1.

17. A system comprising memory which stores instructions for performing the method of claim 1, and a processor in communication with the memory which executes the instructions.

18. A system for recognition of an identifier, comprising:

memory which, for each of a plurality of vehicles, stores;

a first visual signature, which has been extracted from at least a part of a first image of the respective vehicle captured at a first location, and first information associated with the first visual signature, the first information comprising at least one of a first time stamp, a current speed of the vehicle, GPS location, payment information, and weather information, and

a second visual signature, which has been extracted from at least a part of a second image of the respective vehicle captured at a second location, and second information associated with the second visual signature, the second information comprising at least one of a second time stamp, a current speed of the vehicle, GPS location, payment information, and weather information;

a signature matching component for computing a measure of similarity between one of the second visual signatures and at least some of the first visual signatures to identify a matching one of the first visual signatures, based on the computed measure of similarity;

an information processing component which performs a test which is a function of the retrieved stored first information and the stored second information, the first information comprising the at least one of the time stamp, the current speed of the vehicle, the GPS location, the payment information, and the weather information associated with the matching one of the first visual signatures and the stored second information comprising the at least one of the time stamp, the current speed of the vehicle, the GPS location, the payment information, and the weather information associated with the one of the second visual signatures;

an identifier recognition component which recognizes an identifier of one of the plurality of vehicles based on the second captured image of the vehicle when it is confirmed that the test is met; and

a processor which implements the signature matching component, the information processing component, and the identifier recognition component.

19. The system of claim 18, wherein the first information comprises a first time stamp and the second information comprises a second time stamp which is later in time than the first time stamp, the test being a function of the first time stamp and the second time stamp.

20. A method for recognition of a vehicle identifier, comprising:

for each of a plurality of vehicles;

with a first camera, capturing a first image of the vehicle at a first location at a first time,

with a second camera, capturing a second image of the vehicle at a second location at a second time, later than the first time,

acquiring first information associated with the capture of the first image,

acquiring second information associated with the capture of the second image, the first and second information each comprising a respective time stamp,

computing a first visual signature based on at least a part of the first image of the vehicle, the first visual signature being a statistical representation of pixels of the at least a part of the first image,

computing a second visual signature based on at least a part of the second image of the vehicle, the second visual signature being a statistical representation of pixels of the at least a part of the second image,

storing first visual signatures linked to the respective first information, the first visual signatures including the first visual signature for each of the plurality of vehicles, and

storing second visual signatures linked to the respective second information, the second visual signatures including the second visual signature for each of the plurality of vehicles;

computing a measure of similarity between one of the second visual signatures and at least some of the first visual signatures to identify a matching one of the first visual signatures, based on the computed measure of similarity;

retrieving the stored first information associated with the matching one of the first visual signatures and stored second information associated with the one of the second visual signatures;

performing a test which includes computing a function of the time stamps of the retrieved stored first information and the stored second information; and

when it is confirmed that the test is met, recognizing an identifier of one of the vehicles based on the captured first or second image of the vehicle.