Vehicle Side Image Recognition Method And Apparatus

• US 20080158014A1
• Filed: 12/20/2007
• Published: 07/03/2008
• Est. Priority Date: 12/29/2006
• Status: Active Grant

• Current Assignee: Alpine Electronics Incorporated (Alps Alpine Co., Ltd.)
• Original Assignee: Alpine Electronics Incorporated (Alps Alpine Co., Ltd.)
• Inventors: Zou, Bo, Wang, Nan
• Granted Patent: US 8,019,122 B2
• Time in Patent Office: Days
• US Class Current: 340/988
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G06V 2201/08 Detecting or categorising vehicles
• CPC Class Codes: G08G 1/16 Anti-collision systems road vehicle drive control systems for predicting or avoiding probable or impending collision otherwise than by control of a particular sub-unit B60W30/08
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes

Abstract

A vehicle side image recognition method and apparatus inputs images containing recognized vehicles captured by a camera apparatus; sorts images with the recognized vehicles located at left and right sides of the camera apparatus from the input images; performs vehicle side image recognition processing on the sorted images to acquire side image information of each of the recognized vehicles; and outputs the acquired side image information of each of the recognized vehicles. An assisted driving system may use the vehicle side image information to analyze the relative movement of an object vehicle and other vehicles more comprehensively, thus greatly improving the traveling security of a vehicle.

Claims

1. A vehicle side image recognition method, comprising:

inputting images containing recognized vehicles captured by a camera apparatus;

sorting images with the recognized vehicles located at left and right sides of the camera apparatus from the input images;

performing vehicle side image recognition processing on the sorted images to acquire side image information of each of the recognized vehicles; and

outputting the acquired side image information of each of the recognized vehicles.

2. The method according to claim 1, wherein the act of performing vehicle side image recognition processing further comprises:

acquiring a side shadow underneath each of the recognized vehicles from the left and right sides in the sorted images,

acquiring a vehicle side edge of each of the recognized vehicles along the direction of the side shadow underneath each of the recognized vehicles in the sorted images;

determining side image coordinate information of each of the recognized vehicles based on coordinates of the acquired side shadow underneath a vehicle and the acquired vehicle side edge of the recognized vehicle.

3. The method according to claim 2, further comprising:

judging whether an angle difference between the direction of the acquired side shadow underneath a vehicle and that of the acquired vehicle side edge of the recognized vehicle meets a predetermined condition; and

determining the side image coordinate information of the recognized vehicle based on the coordinates of the side shadow underneath the vehicle and the vehicle side edge of the recognized vehicle, if the judgment result is yes.

4. The method according to claim 3, wherein the predetermined condition is that the angle is less than approximately 5 degrees.

5. The method according to claim 2, further comprising:

judging whether a distance between the side shadow underneath a vehicle and the vehicle side edge of the recognized vehicle meets a predetermined condition; and

determining the side image coordinate information of the recognized vehicle based on coordinates of the side shadow underneath the vehicle and the vehicle side edge of the recognized vehicle, if the judgment result is yes.

6. The method according to claim 5, wherein the predetermined condition is that the distance is greater than ⅓

of the height of the recognized vehicle and less than ⅔

of the height of the recognized vehicle.

7. The method according to claim 2, further comprising:

judging whether a lateral distance between the side shadow underneath a vehicle and the vehicle side edge of the recognized vehicle meets a predetermined condition; and

determining the side image coordinate information of the recognized vehicle based on coordinates of the side shadow underneath a vehicle and the vehicle side edge of the recognized vehicle, if the judgment result is yes.

8. The method according to claim 7, wherein the predetermined condition is that the lateral distance is less than 10% of the width of the recognized vehicle.

9. The method according to claim 1, wherein the camera apparatus is mounted on one of an object vehicle or a stationary object.

10. A vehicle side image recognition apparatus, comprising:

an input unit for inputting images containing recognized vehicles captured by a camera apparatus;

a sorting unit for sorting images having recognized vehicles located at left and right sides of the camera apparatus from the input images;

a recognition processing unit for performing vehicle side image recognition processing on the sorted images to acquire side image information of each of the recognized vehicles; and

an output unit for outputting the acquired side image information of each of the recognized vehicles.

11. The apparatus according to claim 10, wherein the recognition processing unit further comprises:

a shadow acquiring module for acquiring a side shadow underneath each of the recognized vehicles from the left and right sides in the sorted images,

an edge acquiring unit for acquiring a vehicle side edge of each of the recognized vehicles along the direction of the side shadow underneath each of the recognized vehicles in the sorted images; and

a determining module for determining side image coordinate information of each of the recognized vehicles based on coordinates of the acquired side shadow underneath the vehicle and the acquired vehicle side edge of the recognized vehicle.

12. The apparatus according to claim 11, wherein the recognition processing unit further comprises a judging module for judging whether an angle difference between the direction of the acquired side shadow underneath a vehicle and that of the acquired vehicle side edge of the recognized vehicle meets a predetermined condition;

and

when the judgment result is yes, the determining module determines the side image coordinate information of the recognized vehicle based on the coordinates of the side shadow underneath the vehicle and the vehicle side edge of the recognized vehicle.

13. The apparatus according to claim 11, wherein the recognition processing unit further comprises a judging module for judging whether a distance between the side shadow underneath a vehicle and the vehicle side edge of the recognized vehicle meets a predetermined condition;

and

when the judgment result is yes, the determining module determines the side image coordinate information of the recognized vehicle based on coordinates of the side shadow underneath the vehicle and the vehicle side edge of the recognized vehicle.

14. The apparatus according to claim 11, wherein the recognition processing unit further comprises a judging module for judging whether a lateral distance between the side shadow underneath a vehicle and the vehicle side edge of the recognized vehicle meets a predetermined condition;

and

when the judgment result is yes, the determining module determines the side image coordinate information of the recognized vehicle based on coordinates of the side shadow underneath the vehicle and the vehicle side edge of the recognized vehicle.

15. A vehicle headlight misrecognition detection method, comprising:

inputting images of recognized vehicle headlights and images containing recognized vehicles captured by a camera apparatus;

acquiring coordinate information of a side image of each of the recognized vehicles based on the input images containing the recognized vehicles;

determining recognized vehicle headlights that are located in side image regions of the recognized vehicles based on the images of the recognized vehicle headlights and the coordinate information of the side image of each of the recognized vehicles; and

outputting the determined recognized vehicle headlights as misrecognized vehicle headlights.

19. The method according to claim 15, wherein the acquiring act further comprises:

sorting images with the recognized vehicles located at left and right sides of the camera apparatus from the input images containing the recognized vehicles; and

performing vehicle side image recognition processing on the sorted images to acquire the coordinate information of the side image of each of the recognized vehicles.

20. The method according to claim 19, wherein the act of performing vehicle side image recognition processing further comprises:

acquiring a side shadow underneath each of the recognized vehicles from the left and right sides in the sorted images,

acquiring a vehicle side edge of each of the recognized vehicles along the direction of the side shadow underneath each of the recognized vehicles in the sorted images; and

determining the side image coordinate information of each of the recognized vehicles based on coordinates of the side shadow underneath the vehicle and the vehicle side edge of the recognized vehicle.

16. A vehicle traveling security prediction method, comprising:

inputting images containing recognized vehicles captured by a camera apparatus mounted on an object vehicle, and a speed of the object vehicle relative to each of the recognized vehicles;

acquiring coordinate information of a side image of each of the recognized vehicle based on the input images containing the recognized vehicles;

calculating an angle of the object vehicle relative to each of recognized vehicles located behind the object vehicle based on the coordinate information of the side image of each of the recognized vehicles;

determining recognized vehicles which are located behind the object vehicle and may collide with the object vehicle based on the calculated angles;

predicting time periods before each of the determined recognized vehicles collide with the object vehicle; and

outputting the determined recognized vehicles and the predicted time periods.

17. The method according to claim 16, wherein the acquiring act further comprises:

sorting images with the recognized vehicles located at left and right sides of the camera apparatus from the input images containing the recognized vehicles; and

performing vehicle side image recognition processing on the sorted images to acquire the coordinate information of the side image of each of the recognized vehicles.

18. The method according to claim 17, wherein the act of performing vehicle side image recognition processing further comprises:

acquiring a side shadow underneath each of the recognized vehicles from the left and right sides in the sorted images,

acquiring a vehicle side edge of each of the recognized vehicles along the direction of the side shadow underneath each of the recognized vehicles in the sorted images; and

determining the side image coordinate information of each of the recognized vehicles based on coordinates of the side shadow underneath the vehicle and the vehicle side edge of the recognized vehicle.