| 摘要 |
<p>PROBLEM TO BE SOLVED: To make an object image included in an original image into parts to be reused easily. SOLUTION: An original image is read through reading a color original by an image reader (100), and discontinuous information is detected by a line process (102). Discontinuous parts in the image are extracted as the discontinuous information and parts formed by linking the discontinuous parts are extracted as a contour of the object image from the fact that the line process is turned on or off, depending on local density difference. Next, color areas with similar colors in the original image are set (104), and the original image is classified (106). The object image included in the classified image is extracted, and the discontinuous parts (contour) are at the same time outlined by each classified image (108 to 114). The object image is made into parts by describing the object image by the outlined contour and color information of the object image (116 to 120).</p> |
