A motion compression method by fuzzy relational equations
Hajime Nobuhara, Kaoru Hirota · 2004
A motion compression method by max t-norm composite fuzzy relational equations (MCF) is proposed. In the case of MCF, a motion sequence is divided into intra-pictures (I-pictures) and predictive-pictures (P-pictures). The I-pictures and the P-pictures are compressed by using uniform coders and non-uniform coders, respectively. In order to perform an effective compression and reconstruction of the P-pictures, a design method of non-uniform coders is proposed. The proposed design method is based on an overlap level of fuzzy sets and a fuzzy equalization. Through a experiment using 10 P-pictures, it is confirmed that the root means square errors of the proposed method is decreased to 89.4% of the uniform coders, under the condition that the compression rate is 0.0057. A result of motion compression and reconstruction of motion (standard motion imagery) by MCF is also presented.