Quadtree-based Dot Diffusion Block Truncation Coding

Abstract

　　In the age of information explosion reducing the storage and bandwidth has become an important issue in order to store and transmit images efficiently Block Truncation Coding (BTC) is a still image compression technique with the feature of simple and fast However the bit rate of the BTC algorithm is relatively high compared to modern compression techniques such as JPEG or JPEG2000 At present modified and improved BTC algorithms are proposed but these algorithms produce block effect and noise while trying to promote compression ratio Therefore a low-complexity BTC technique that can reduce the perceptual artifacts effectively and provide good image quality at low bit rate is essential 　　In this thesis the proposed method is improved based on the halftoning-based BTC Quadtree decomposition method segments the larger non-overlapping blocks of an image into smaller blocks based on the content of the image To reduce the problem of high compression ratio dot diffusion technique which diffuses the quantized error into neighboring unprocessed pixels is adopted to maintain the local gray level 　　According to the smoothness of different blocks processes the experiment result shows that our proposed method has better performance in reducing the bit rate of either image quality assessment or perceptual observation

Date of Award	2014 Jul 21
Original language	English
Supervisor	Pei-Yin Chen (Supervisor)