| 摘要 |
<P>PROBLEM TO BE SOLVED: To efficiently extract characters and /or figures from among digital images. <P>SOLUTION: As shown in Fig. 4, when a digital image 111 is a general color image, a vision apparatus 2 classifies the digital image 111 into at least five colors, i.e., white, black, red, green, and gray, using a polychromatic classification means 11 to enable the generation of a polychromatic image. Thereby, the visual apparatus 2 can compress a single pixel of the digital image 111 into 3 bits. Further, as shown in Fig. 5, the visual apparatus 2 extracts the characters and/or the figures and these contours from the polychromatic image using a character/figure selecting means 12 and also eliminates textures using a texture eliminating means 13. As a result, the visual apparatus 2 can efficiently extract the characters and /or the figures in the digital image 111. <P>COPYRIGHT: (C)2005,JPO&NCIPI |
