Affiliation:
1. East China Jiao Tong University
Abstract
The adaptive region algorithm is an improved compression algorithm based on Huffman
coding. Because of the large number of rules for dividing regions in
the algorithm, there are problems such as high computing costs, slow
speed, and low compression efficiency. To address these problems, this
paper investigates the adaptive region algorithm based on a ternary
optical computer (TOC) combined with the characteristics of a ternary
optical computer, such as many data bits, high parallelism, and
three-valued coding. According to the characteristics of TOC
three-valued coding, this paper designs a three-valued character
coding scheme that can effectively shorten the coding length of
characters by changing the original coding rules and further improve
the compression efficiency of the adaptive region algorithm.
Furthermore, in conjunction with the advantages of TOC enabling
parallel computation, this paper presents an efficient computational
scheme capable of effectively improving computational efficiency
during the process of region partitioning. Through case studies, the
compression efficiency and computational efficiency of the adaptive
region algorithm implemented on TOC and an electronic computer were
analyzed, respectively. It was found that the compression efficiency
of the TOC-based algorithm is 50.4%, while that of the
electronic-computer-based algorithm is only 36%. In the comparison of
computational efficiency, the computational time complexity of TOC is O(n), whereas that of the electronic
computer (EC) is O(n2). Finally, it is concluded through
experimental validation that the TOC-based adaptive region compression
algorithm performs well in terms of computational performance and
compression efficiency, giving full play to the three-valued coding
characteristics of TOC-based as well as the advantages of being able
to realize parallel computation.
Funder
Natural Science Foundation of Jiangxi Province
National Natural Science Foundation of China