To easily correct the original image inputted from various input equipment without using a prescribed parameter by adapting the black character color of the output image to the original image.
Character regions are extracted from the original image in step S5 on the basis of a feature calculated from the original image as a digital image in step S4, the original image is divided into blocks in step S6, and pixels belonging to the extracted character regions are classified into first and second classes according to the colors in step S7. Then, the color tone of the black characters is corrected in step S9 based on the color of the black characters on the original image estimated for each block in step S8 on the basis of the classification of the pixels belonging to the extracted character regions.
COPYRIGHT: (C)2004,JPO
JP8307666A | ||||
JP2001169080A | ||||
JP2001223915A |