METHOD FOR READING VEHICLE IDENTIFICATIONS

• US 20140270386A1
• Filed: 03/12/2014
• Published: 09/18/2014
• Est. Priority Date: 03/13/2013
• Status: Active Grant

• Current Assignee: Kapsch TrafficCom AG (Datax Handelsgmbh)
• Original Assignee: Kapsch TrafficCom AG (Datax Handelsgmbh)
• Inventors: Leihs, Dietrich, Abl, Alexander, Nagy, Oliver
• Granted Patent: US 9,361,535 B2
• Time in Patent Office: Days
• US Class Current: 382/105
• 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 20/63 Scene text, e.g. street names
• CPC Class Codes: G08G 1/0175 by photographing vehicles, e.g. when violating traffic rules

Abstract

A method is disclosed for reading license plate numbers in a road network, comprising: recording an image of a license plate number at a first location, OCR-reading a license plate character string in the image, and storing an OCR data set in a database; recording an image of a license plate number at a second location, OCR-reading a license plate number character string in the image, and generating a current OCR data set; and, if a confidence measure of the current OCR data set falls below a first minimum confidence value, selecting a stored OCR data set with a license plate number image having a similarity exceeding a minimum similarity value and/or having the greatest similarity to the license plate number image of the current OCR data set, and using the selected OCR data set for improving the license plate number character string of the current OCR data set.

Claims

1. A method for reading license plate numbers by means of optical character recognition (OCR) in a road network, comprising:

in a first OCR reading process, recording a first image of a license plate number at a first location in the road network, OCR-reading a first license plate character string in the first image recording, and storing a first OCR data set including the first license plate number image recording, the first license plate number character string and at least one first confidence measure of the first OCR reading process, as a stored OCR data set in a database;

in a second OCR reading process, recording a second image of a license plate number at a second location in the road network, OCR-reading a second license plate number character string in the second image recording, and generating a second OCR data set including the second license plate number image recording, the second license plate number character string and at least one second confidence measure of the second OCR reading process, as a current OCR data set;

and, if at least one confidence measure of the current OCR data set falls below a first minimum confidence value, selecting at least one stored OCR data set from the database, the license plate number image recording of which selected stored OCR data set has a similarity that exceeds a minimum similarity value and/or has the greatest respective similarity to the license plate number image recording of the current OCR data set, and using the at least one selected OCR data set for improving the license plate number character string of the current OCR data set.

2. The method according to claim 1, wherein based on a plurality of different first locations, a plurality of different first OCR data sets are stored in the database.

3. The method according to claim 1, wherein the license plate number character string of the current OCR data set is used for tolling a location usage of a vehicle in the road network.

4. The method according to claim 1, wherein said improving includes superposing the license plate number image recording(s) of the at least one selected stored OCR data set and the license plate number image recording of the current OCR data set, and by OCR-reading again the license plate number character string of the current OCR data set in the superposition.

5. The method according to claim 1, wherein said improving includes taking over the license plate number character string of the selected stored OCR data set into the current OCR data set if at least one confidence measure of the selected stored OCR data set exceeds a second minimum confidence value and/or is greater than the corresponding confidence measure of the current OCR data set.

6. The method according to claim 1, wherein the first OCR data set is stored as stored OCR data set in the database only if the at least one confidence measure of the first OCR data set exceeds a third minimum confidence value.

7. The method according to claim 1, wherein during one or more of the first and second OCR reading processes, a separate confidence measure is generated for each individual character of the character string.

8. The method according to claim 7, wherein another confidence measure is averaged from the confidence measures of all characters of the character string.

9. The method according to claim 1, wherein each stored OCR data set and the current OCR data set also contain in each case the location of their license plate number image recording, and that during said selecting only those stored OCR data sets are considered, the locations of which lie in a predetermined region around the location of the current OCR data set.

10. The method according to claim 1, wherein each stored OCR data set and the current OCR data set also contain in each case the time of their license plate number image recording, and that during said selecting only those stored OCR data sets are considered, the times of which lie within a predetermined period around the time of the current OCR data set.

11. The method according to claim 1, wherein the first OCR data set is stored as stored OCR data set in the database only for a predetermined period.

12. The method according to claim 2, wherein the first locations comprise some or all intersections of the road network with a closed geographical region.

13. The method according to claim 12, wherein the geographical region is a city or a country.

14. The method according to claim 1, wherein recording an image at at least one first or second location is carried out from a moving control vehicle.

15. The method according to claim 2, wherein the database is kept in a control center of the road network, and the first OCR data sets to be stored are transmitted from the first locations to the control center, and the current OCR data set is transmitted from the second location to the control center for said selecting.

16. The method according to claim 2, wherein the database is kept locally at one of the first locations, and the current OCR data set is transmitted from the second location to a nearest first location for said selecting.