Edge-Oriented Image Compression Technique using Edge Prediction and Classified Vector Quantization
Abstract
Vector quantization (VQ) has been extensively employed for image compression. In this paper, a predictive classified vector quantization (PCVQ) algorithm in the discrete cosine transform (DCT) domain is presented. A total of 221 subcodebooks are generated by exploiting the edge information for 8x8 blocks which affects the distribution of DCT coefficients. The encoding complexity is reduced significantly since the codebook is divided into subcodebooks with much smaller sizes and dimensions. Simulation results show that a comparable reconstruction quality is obtained for a given bit rate with a significant reduction in complexity. A simple and efficient classification algorithm is developed by using two DCT co-efficients and spatial data. The performance of the proposed DCT-PCVQ coding technique was evaluated using two standard test images. The results obtained are comparable to the best results reported so far for the same set of images tested here.
Recommended Citation
C. K. Chow and F. Ercal, "Edge-Oriented Image Compression Technique using Edge Prediction and Classified Vector Quantization," Proceedings of the Data Compression Conference, p. 478, Jan 1994.
Department(s)
Computer Science
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2024 The Authors, All rights reserved.
Publication Date
01 Jan 1994