| 摘要 |
<P>PROBLEM TO BE SOLVED: To improve a method for compressing a composite image including an image and a text. <P>SOLUTION: Multiple resolutions are used in segmentation/fill techniques. In one embodiment of the present invention, the method comprises generating a plurality of layers using image data from the multiple resolutions, generating a mask describing composing of the plurality of layers to obtain the image, and filling pixels in each of the plurality of layers where the pixels being filled are those that do not contribute to the image when the plurality of layers are composed. <P>COPYRIGHT: (C)2006,JPO&NCIPI |
