抄録
In the computer vision, it is important problem to recover a three-dimensional object from a two-dimensional image. Because, in the industry field, it is necessary to develop a vision system to improve productivity and efficiency. But this problem is very difficult and there are various devices were studied. However there are some problems using these devices, it costs much time to measure the 3D shape, the resolution of shape data is low, and the devices are very expensive. In this paper, to improve such problems, we propose the method to recover the 3D shape of an object using a color image scanner which has three light sources. By using the image scanner, the 3D shape data will be able to get more easily. Hence, our final goal is to change the image scanner into the shape scanner.
