| 摘要 |
PROBLEM TO BE SOLVED: To provide "an obstacle detecting apparatus and an obstacle detecting method", reducing a processing load for analyzing an image executed for detecting an obstacle, and saving a memory source, without worsening detection precision for the obstacle as much as possible. SOLUTION: A motion vector is detected with the reduced processing load and the reduced use of the memory source, using only one part of an area specified out of the whole area as an object, by partitioning the whole area of monitoring range in the image of each frame photographed by an onboard camera 20 into a plurality of small areas, and by comparing the image of the current frame with the image of the preceding frame, using, as the objects, the images of the plurality of small areas (sections) existing in a specified position in the whole area, to detect the motion vector of the obstacle, and trouble of thinning completely the frame photographed with the obstacle is eliminated to restrain, to the utmost, the motion vector from being not detected even in the situation of the presence of the obstacle. COPYRIGHT: (C)2011,JPO&INPIT |
