A digital sharpening algorithm which operates in the luminance, hue and saturation (LHS) color space was applied to simulated images and digital images scanned on a CMYK scanner. The spatial algorithm was the same as the one used in normal operations. Edge detection was performed in the HS space on the luminance (L) image and the sharp signal derived from it was added to the original unsharpened L image. The effect of sharpening the saturation (S) image was also tested. The resulting images were transformed back to the CMYK space, printed and compared to images sharpened in the RGB space with various conventional algorithms. Results show that sharpening in the LHS space avoids contours with color different from that of the object and maintains the original color of very fine details, in contrast to the conventional RGB sharpening algorithm. Some aspects of this method, however, should still be improved before its advantages can justify the changes necessary for implementation in the color separation scanner.