A New Low Bit-Rate Image Coding Scheme Based On Human Visual System

Image compression has been research in information processing technology which is more active position, it's the key technologies of the multimedia computer, data distribution, modern communication and media storage. In order to improve the efficiency of the image information transmission, computer processing speed and storage capacity, a large number of effective data storage, processing and transmission of multimedia information processing in the field of digital image compression coding technology status has been increasingly more important. This paper is a study in this background.In the field of image coding, the study's core technology and hotspots is a multi-scale geometric analysis theory, the wavelet analysis in multiscale geometric analysis tool has been made in the classic and very good compression. However, due to the contour, the edge of high-dimensional data information seems powerless to overcome this deficiency, there are many new tools in recent years, multi-scale analysis based on wavelet transform generated, Contourlet transform in this case is to be transported and students. Compared with the wavelet transform, Contourlet transform not only to efficiently capture the image geometry, and anisotropy and has a good direction.Firstly, in this paper from the perspective of transform coding, introduced the theorety and Research of the wavelet transform and transform and. Secondly, the focus of this study is based on the lifting wavelet transform and transform, Binding experiments to study the two transform domain image compression algorithm: A: A New Low Bit Rate Image Coding Scheme Based On Human Visual System. Firstly, the original image using lifting wavelet decomposition, the visual gains are achieved by exploiting HVS characteristics to weigh wavelet coefficients according to their perceptual importance. Lastly, the lower Frequency exploit a new partitioning structure based on the importance of coefficients. and completed by the idea of the SPIHT image compression coding. B:A New Image Coding Scheme Based On Transform and Human Visual System. At the first images using wavelet transform decomposition, and further implementation of the adaptive directional wavelet decomposition in the high-frequency sub-band; then human visual system (HVS), to the rules given by different visual sub-blocks of different weights; then based on Wavelet Transform and transform the characteristics of the spatial orientation tree structure to re-structure; Finally, according to the importance factor of a new arrangement of low-frequency coefficients of the structure, and using SPIHT coding idea of the image compression. The experimental results show that the proposed two algorithms are better than generally applicable, the classic image coding (especially in low bit rate), with relatively strong adaptability and versatility, it has the good application.