| 摘要 |
PROBLEM TO BE SOLVED: To perform the image processing operation of an object to be recognized, regarding various optical characteristics by a method wherein the object to be recognized is recognized on the basis of a plurality of images photographed by a high-speed photographing camera by changing an illumination condition when it is photographed. SOLUTION: When a first image is photographed, first illumination devices 3 are turned on by a controller 7, and the image which is photographed by a double-speed photographing camera 5 is stored in an image memory at an image processor 6 via an exposure operation A and a transfer operation A. Then, simultaneously with the finish of the exposure operation A, the illumination devices are changed over to second illumination devices 4 by the controller 7, and the photographing operation of a second image is started, and, after that, the same procedure is performed as when the first image is photographed. The image processor 6 selects the image, which is optimum for a successive image processing operation, out of the two images stored in the image memory, and it recognizes the position and the posture of an object 2 to be recognized by the image processing operation. The starting point of the image processing operation at the image processor 6 in a point of time in which the first image is stored in the image memory, i.e., a point of time in which the transfer operation A is finished, the image processing operation is executed in parallel with the transfer operation B of the second image, and the processing time is shortened. |
