IMAGE TYPE DISCRIMINATION DEVICE, IMAGE PROCESSOR USING THE SAME, AND IMAGE TYPE DISCRIMINATION METHOD

摘要

PROBLEM TO BE SOLVED: To automatically discriminate an artificial image or a natural image per pixel to select proper image processing. SOLUTION: A delay difference by one line is given to image data, which is inputted into a first correlation calculation part 730a, and 3×3 vicinity pixel is sampled for a judging pixel whose type of image is determined. The number of vicinity pixels in which the determined pixel agrees with the image data is counted, and the counted value is compared with a value of a parameter 1 of a register 728 to perform primary determination of type of image. The results of determination to be given to each pixel are further supplied to a second correlation calculation part 760a to sample 3×3 vicinity pixel for the determined pixel in the same way. The number of types of images in the results of primary determination is counted, and the counted value is compared with a parameter 2 with respect to size to discriminate type of image of the pixel concerning the vicinity pixels in a 3×3 region centered on the determined pixel. The optimum processing is selected in interpolation processing and emphasis processing using the results of determination.