LICENSE PLATE MATCHING SYSTEMS AND METHODS

• US 20160299897A1
• Filed: 04/08/2016
• Published: 10/13/2016
• Est. Priority Date: 04/09/2015
• Status: Active Grant

• Current Assignee: VeriToll LLC
• Original Assignee: VeriToll LLC
• Inventors: Silva, Joseph C., Higgins, Christopher T.
• Granted Patent: US 10,901,967 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2228 Indexing structures
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06V 20/54 of traffic, e.g. cars on the road, trains or boats
• CPC Class Codes: G06V 20/625 License plates

Abstract

Systems and methods for license plate matching. In one example, a system for license plate matching includes an interface configured to receive roadside data, the roadside data including a roadside image of a vehicle license plate, a database having a plurality of indexed entries, each entry including vehicle license plate fingerprint data and vehicle characteristic metadata, and a processor connected to a memory, the interface, and the database, the processor configured to execute a query against the database based on the roadside data, compare a portion of the entries to the roadside data, the processor being further configured to compare the roadside image of the vehicle license plate and the corresponding vehicle license plate fingerprint data when comparing the portion of the entries and roadside data, and generate a confidence based on a similarity between at least one individual entry of the portion of entries and the roadside data.

Claims

1. A system for license plate matching comprising:

a system interface configured to receive roadside data from a remote device, the roadside data including at least a roadside image of a first vehicle license plate;

a database having a plurality of indexed entries, each entry of the plurality of entries including at least vehicle license plate fingerprint data and vehicle characteristic metadata; and

at least one processor operatively connected to a memory, the system interface, and the database, the processor when executing configured to;

execute a query against the database based on the roadside data,

compare at least a portion of the entries to the roadside data, the processor when executing being further configured to compare the roadside image of the first vehicle license plate and at least the corresponding vehicle license plate fingerprint data when comparing the portion of the entries and the roadside data, and

generate a confidence factor based in part on a degree of similarity between at least one individual entry of the portion of the entries and the roadside data.

2. The system for license plate matching according to claim 1, wherein the vehicle license plate fingerprint data includes image data extracted from a dynamically generated composite image of a virtual vehicle license plate.

3. The system for license plate matching according to claim 2, wherein the virtual vehicle license plate includes at least a predetermined series of characters.

4. The system for license plate matching according to claim 3, wherein the first vehicle license plate includes at least a first series of characters, and wherein in comparing the roadside image of the first vehicle license plate and the vehicle license plate fingerprint data, the processor when executing is further configured to compare the vehicle license plate fingerprint data and the roadside image of the first vehicle license plate so as to recognize at least one of the first series of characters.

5. The system for license plate matching according to claim 4, wherein the first series of characters includes a vehicle registration identifier.

6. The system for license plate matching according to claim 1, wherein the roadside data further includes roadside vehicle characteristic metadata, and wherein in comparing the portion of the entries to the roadside data, the processor when executing is further configured to compare the roadside vehicle characteristic metadata and at least the corresponding vehicle characteristic metadata of the portion of the entries.

7. The system for license plate matching according to claim 6, wherein the vehicle characteristic metadata includes at least one of speed, color, size, classification, age, tire noise, metal content, axes count, tire count, aerodynamic characteristics, engine noise, vehicle noise, and spatial contours.

8. The system for license plate matching according to claim 1, wherein the processor when executing is further configured to compare the confidence factor to a confidence threshold.

9. The system for license plate matching according to claim 8, wherein in response to a determination that the confidence factor is greater than the confidence threshold, the processor when executing is further configured to generate a new indexed entry based at least in part on the roadside data.

10. The system for license plate matching according to claim 1, wherein the system interface is further configured to receive the roadside data from a user device.

11. The system for license plate matching according to claim 1, wherein the database includes one or more cloud database instances provisioned among one or more cloud database providers.

12. The system for license plate matching according to claim 1, wherein the vehicle license fingerprint data of each entry includes at least one binary index.

13. A method for license plate matching, the method comprising:

receiving roadside data including at least a roadside image of a first vehicle license plate;

querying a database having a plurality of indexed entries, each entry of the plurality of entries within the database including at least vehicle license plate fingerprint data and vehicle characteristic metadata;

comparing at least a portion of the entries to the roadside data, including comparing the roadside image of the first vehicle license plate and at least the corresponding vehicle license plate fingerprint data; and

generating a confidence factor based in part on a level of similarity between at least one individual entry of the portion of the entries and the roadside data.

14. The method according to claim 13, the method further comprising dynamically generating a composite image of a virtual vehicle license plate, and wherein the vehicle license plate fingerprint data of at least one entry includes image data extracted from the composite image.

15. The method according to claim 14, wherein comparing the roadside image of the first vehicle license plate and the vehicle license plate fingerprint data further includes comparing the vehicle license plate fingerprint data and the roadside image of the first vehicle license plate so as to recognize at least one of a first series of characters of the first vehicle license plate.

16. The method according to claim 13, wherein comparing the portion of the entries to the roadside data includes comparing the roadside vehicle characteristic metadata and at least the corresponding vehicle characteristic metadata.

17. The method according to claim 13, the method further comprising comparing the confidence factor to a confidence threshold.

18. The method according to claim 17, the method further comprising generating a new entry based at least in part on the roadside data and responsive to determining that the confidence factor is greater than the confidence threshold.

19. The method according to claim 13, wherein querying the database further includes querying one or more cloud compute database instances provisioned among one or more cloud database providers.

20. A method for vehicle license plate indexing, the method comprising:

receiving vehicle license plate metadata;

dynamically generating a composite image of a virtual vehicle license plate based on the received vehicle license plate metadata;

generating vehicle license plate fingerprint data, the vehicle license plate fingerprint data including image data extracted from the composite image; and

generating an indexed entry for a vehicle license plate matching system database based at least in part on the vehicle license plate fingerprint data.