| 摘要 |
Color space conversion and correction processing are performed on inputted R, G and B image data 1 by an RF converter 2, and Y, M, C and K image data 3 is obtained. In the case where Y, M, C and K or Y, M and C image data is directly inputted, the RF converter 2 is passed through without being processed. Inverse processing is performed on the Y, M, C and K image data 3 by a conversion table 6 where a conversion table is set in advance, and the original R, G and B image data is obtained. An image recognition circuit 7 recognizes an image on the basis of the R, G and B image data outputted from the conversion table 6. By virtue of the above process, when an image of a specified document is to be recognized, image recognition can be performed with high precision on the basis of R, G and B image data where inverse conversion has been performed on Y, M, C and K image data or Y, M and C image data.
|
