| 摘要 |
PROBLEM TO BE SOLVED: To solve the problem that it is difficult to improve a recognition rate and reduce erroneous recognition in a system for recognizing a target object around a vehicle by an on-vehicle camera, wherein only a rear camera is used for recognition conventionally, so that it is difficult to predict where a road marking appears in a screen, which appears from a lower end of the screen in the image of the rear camera, and a target can be captured only for a short time because the camera is set with a large depression angle. SOLUTION: The result of recognizing the target object (the kind, position, angle, recognition time of the object) by the camera installed at the front of a vehicle is used to predict the appearance timing and position of the target object in the visual field of the vehicle rear camera, and the parameter and processing timing of recognition logic of the vehicle rear camera are optimally adjusted. Further, luminance information of an image of the camera installed at the front of the vehicle is used to predict change of the luminance of the visual field of the vehicle rear camera, and the gain and exposure time of the vehicle rear camera are adjusted. COPYRIGHT: (C)2008,JPO&INPIT
|
