Video preprocessing of image information for character recognition

• US 8,218,821 B2
• Filed: 01/23/2007
• Issued: 07/10/2012
• Est. Priority Date: 01/23/2007
• Status: Active Grant

• Current Assignee: Neology Incorporated (OEP Technologie BV)
• Original Assignee: PIPS Technology Incorporated (3M Company)
• Inventors: Sefton, Alan K., Rinehart, Kent A.
• Primary Examiner(s): Harold, Jefferey
• Assistant Examiner(s): Khalid, Omer
• Application Number:
  US11/625,896
• Publication Number:
  US 20080175479A1
• Time in Patent Office: 1,995 Days
• Field of Search: 386/242, 725/12, 902/6, 382181-182, 382/105, 358/909.1, 362/497, 402/00
• US Class Current: 382/105
• CPC Class Codes: G06V 20/625 License plates
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names

Abstract

An apparatus processes video signals containing video information related to a scene which may contain a vehicle license plate. The apparatus includes a video camera having a video imaging device for viewing the scene and generating a first video signal. A character detector in the video camera processes the first video signal to detect a license plate within the scene and to generate location information indicating the location of the license plate. A line detector in the camera determines a particular video line of the first video signal into which the location information is to be embedded. An insertion circuit in the camera embeds the location information into the particular video line of the first video signal to form a second video signal. The apparatus may also include a video capture device for receiving the second video signal from the video camera and converting the second video signal into digital image data. A processor extracts the location information from the digital image data, determines the location of the license plate characters within the scene based on the location information, and processes the digital image data to read the license plate characters.

Claims

1. A method for processing video information related to a scene which may contain characters of interest, the method comprising:

(a) acquiring a video signal comprising a plurality of horizontal lines containing the video information related to the scene;

(b) processing the video signal to detect whether characters of interest are present in the scene;

(c) embedding indicator information into one or more of the horizontal lines of the video signal, where the indicator information indicates the presence of the characters of interest in the scene;

(d) converting the video signal with the embedded indicator information into digital image data representing the scene;

(e) extracting the indicator information from the digital image data;

(f) determining based on the indicator information that characters of interest are present in the scene; and

(g) processing the digital image data to determine the characters of interest.

2. The method of claim 1 wherein:

step (b) comprises processing the video signal to determine location information indicating a location of the characters of interest within the scene;

step (c) comprises embedding the location information into one or more of the horizontal lines of the video signal;

step (e) comprises extracting the location information from the digital image data; and

step (g) comprises processing the digital image data based at least in part on the location information.

3. The method of claim 1 wherein step (c) includes embedding the indicator information into a particular video line of one or more video frames in the video signal.

4. The method of claim 1 wherein step (c) includes embedding the indicator information in a bit-serialized format in the video signal.

5. The method of claim 1 wherein the characters of interest comprise characters of a vehicle license plate.

6. The method of claim 1 wherein step (g) comprises processing the digital image data only if the indicator information indicates that characters of interest are present within the scene.

7. The method of claim 1 wherein step (a) comprises acquiring the video signal from a video camera viewing the scene.

8. A method for processing a video signal comprising a plurality of horizontal lines containing video information related to a scene containing characters of interest, the method comprising:

(a) acquiring the video signal from a video camera viewing the scene;

(b) processing one or more frames of the video signal to determine a location of characters of interest within the scene;

(c) generating location coordinates indicating the location of the characters of interest within the scene, wherein the location coordinates are values referenced to a top or bottom of the one or more frames, or are X-Y coordinate values; and

(d) embedding the location coordinates in one or more of the horizontal lines of the video signal.

9. An apparatus for processing video information related to a scene which may contain characters of interest, the apparatus including a video camera comprising:

a video imaging device for viewing the scene and generating a first video signal comprising a plurality of horizontal lines;

a character detector for processing the first video signal to detect characters of interest present in the scene;

an insertion circuit for embedding indicator information within one or more of the horizontal lines of the first video signal to form a second video signal, the indicator information indicating that the characters of interest are present in the scene;

a camera housing for enclosing the video imaging device, the character detector and the insertion circuit; and

a video output connector on the camera housing for providing electrical connection to the second video signal.

10. The apparatus of claim 9 further comprising a line detector for determining a video line of one or more video frames of the first video signal into which the indicator information is to be embedded.

11. The apparatus of claim 9 further comprising:

a video capture device for receiving the second video signal from the video camera and converting the second video signal into digital image data representing the scene; and

a processor for extracting the indicator information from the digital image data, for determining based on the indicator information whether the characters of interest are present in the scene, and for processing the digital image data to read the characters of interest.

12. The apparatus of claim 11 wherein:

the character detector processes the first video signal to determine location information indicating a location of the characters of interest within the scene;

the insertion circuit embeds the location information within one or more of the horizontal lines of the first video signal to form the second video signal; and

the processor extracts the location information from the digital image data and processes the digital image data based at least in part on the location information.

13. The apparatus of claim 9 wherein the insertion circuit embeds the indicator information within a particular video line of one or more video frames in the first video signal to form the second video signal.

14. The apparatus of claim 9 wherein the insertion circuit embeds the indicator information in a bit-serialized format within the first video signal to form the second video signal.