Study On Digital Image Compression Technique

Digital Library is an information resource management mode adopted by the coming generation super high speed Internet. During its building, data compression technology plays a more and more important role. However, the current transmission and storage capability requires more for the low bit rate narrow communication. How to compress data efficiently is still a hotspot. This paper studies the image compression technology in digital library information resources construction.This paper mainly discusses the hybrid-coding algorithm of static image. Based on analyzing the advantage and disadvantage of the existing usual compression methods, Self-organizing feature mapping neural network is used to carry out adaptive scalar quantization, at the same time, after wavelet transformation, different resolution quantization is applied to different visual sensitive frequency section to acquire good compression effect.In addition, this paper presents three kinds of compression methods for color image, which includes the method using wavelet transformation to realize color difference and luminosity spatial resolution compression; the method using different quantization resolution and the method using SOFM to realize indexed color transformation compression. The experimental results show the above methods can all achieve good compression effect and high compression ratio.