Appeal No. 1999-2040 Application No. 08/475,023 Page 18 correction matrix calculation (col. 2, lines 35-43). As shown in figure 1, a human subject image is taken by a color TV camera. A/D converter 14 converts the R, G, B signals to digital form (col. 3, lines 25-40). CG images are inputted from diskette 17 (col. 3, lines 43-45). Color correction circuit 21 performs color correction of human subject images, CG images, and character images using a color correction matrix (col. 4, lines 3-6). As shown in figure 3, table data 50a representing pure color R,G,B CG images are not corrected in gradation, and human subject image table data 51a is non-linear. As shown in figure 7, compound color CG images have a bent-line table (col. 5, lines 52-62). Characters such as names are entered through the keyboard and are subjected to the same image processing as CG images (col. 8, lines 7-11). The gradation-corrected image signals are transmitted to color-correction circuit 21 (col. 7, lines 41 and 42). Figure 8 displays a printed identification card having the composite image. In addition, Oshikoshi discloses that the human subject image and the CG image can be composed "at one of before and after" correction of gradation (col. 9, lines 44-48). We note that appellants do not provide any specific reasons to support their conclusionary statement (brief, page 13) thatPage: Previous 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NextLast modified: November 3, 2007