| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To enhance accuracy in determining whether a processing object image is a graphic. <P>SOLUTION: The processing object image is divided into color regions (clusters), information (number of labels, number of pixels), with respect to the connection region (a label region) in each of the divided color regions is acquired. Based on the information acquired, the processing object image is determined as to whether if it is graphics. In this fashion, determination is made with higher accuracy as to it is a graphic that is appropriate to vectorization processings. <P>COPYRIGHT: (C)2010,JPO&INPIT</p> |
