| 摘要 |
A vision method for automatically detecting objects in front of a motor vehicle, comprises the steps of detecting images from a region in front of the vehicle by a vehicle mounted imaging means (11), generating a processed image (23), containing disparity or vehicle-to-scene distance information, from said detected images, comparing regions-of-interest of said processed image (23) to template objects (24A, 24B, 24C, ...) relating to possible objects (21a, 21b) in front of the motor vehicle, calculating for each template object (24A, 24B, 24C, ...) a score relating to the match between said processed image (23) and the template object (24A, 24B, 24C, ...), identifying large scores (31A, 31B, 31C, ...) of sufficient absolute magnitude of said calculated scores. The vision method comprises the steps of applying an algorithm adapted to identify groups (31A-31C; 31D-31I) of said large scores, and assigning an identified group (31A-31C; 31D-31I) of large scores to a single match object. |
