作者: S. Vimala , K. Somasundaram
DOI:
关键词: Code (cryptography) 、 Algorithm 、 Mathematics 、 Computation 、 Sorting 、 Distortion 、 Codebook 、 Feature (computer vision) 、 Linde–Buzo–Gray algorithm 、 Vector quantization
摘要: In this paper, we present a new and fast encoding algorithm (FEA) for vector quantization. The magnitude (sum of the components vector) feature vectors is used in to improve efficiency searching. Sorting values enhances As are sorted, searching can be terminated advance reduce time needed locate representative code vector. For codebook size M (M generally being 128/256/512/1024), distortion calculations performed. But proposed method, only 11 computations done irrespective codebook. taken codevector significantly reduced from 0.77 seconds 0.07 on an average. experiments were carried over with codebooks sizes 128, 256, 512 1024 standard images Lena, Boats, Cameraman Bridge.