Codeword Mapping Optimization Method for Minimization of Recognized False Contour Noise

Abstract

A new codeword mapping optimization method is proposed to reduce the recognized amount of false contour noise (FCN). The proposed method is based on the fact that the FCN generated at the object surface is perceived more significantly than the FCN generated at the object edge and minimizes object surface FCN first by mapping similar codeword intensively to gray levels consisting of that surface. For implementation, a new FCN evaluation method which considers the FCNs generated in all possible gray level pairs is introduced and the cost function and the search space of the new codeword mapping optimization method are derived from the FCN evaluation method. Finally, a method which calculates the recognized amount of FCN from the generated amount of FCN is introduced(1).