摘要近期，数字图像技术飞速发展，这极大的促使了多媒体娱乐、通信和高清电视等各类与图片或者视频有关系的产品们的发展。同时，图像信息被几乎所有人应用于多媒体以及计算机系统中，然而图像数据的最为明显的特点就是信息量超级的大，不不仅仅通过计算机难以完成，而且以如今通信的信道传输速率，是没有办法完成大数据量多媒体信息的传输的。86673

图像压缩是人们在数字图像上的应用，其目的是为了可以减少图像数据中的非常多的多余信息从而得以用更高效的格式来进行存储和传输。所以保证图像质量的条件下，用尽可能少的比特数来提传输图像时的效率并且减少存储图像所需占用的容量。

在此，本文选择了用以下两个较为常用的方法来对图像压缩：

 1。基于小波变换的图像压缩：小波变换压缩图像的特点就是压缩比高，数字图像和信号的各方面因素可以在压缩后依旧不变，同时在信号传递中，它也可以抗干扰。其原理就是将图像信号来进行域的转换，变换图像的主体部分几乎都集中在很少的变换系数部分，并且对这些变换系数进行量化编码来进行压缩。

 2。基于DCT的图像压缩：DCT同时也叫离散余弦变换是通过傅立叶变换性质，把图像边界的褶翻来从而将图像变为偶函数形式，来进行二维傅立叶的变换，并且留下余弦项，来把图像中的空间冗余去掉来压缩。

毕业论文关键词：图像压缩；MATLAB；小波；DCT

Abstract  In recent years, image technology develops rapidly and at the same time it pushes the development of multimedia, multimedia-communication, high definition television and all kinds of relative image and video production。 At the same time, image information is used in multimedia communication and computer system widely。 However, one typical characteristic of image data is large content which not only beyond the reserve and manage ability of computer but also can not complete a large quantity of multimedia communication live transmission under the speed of existing communication channel。

Image compression is an application of data compression technology on digital image, which goal is to format reservation and transmit data with high efficient by reduce the redundant information from image data。 Under the requirement of image quality, using less bite to improve the efficiency of image transmission and to reduce the capacity of image reservation。

This research will use two approaches to compress image:

1。The image compression under wavelet transform: which characteristic is high compression rate and the feature of image and signal can remain unchanged after compressing and has high anti-interference ability of signal transmission。 The basic principal of wavelet transform in image compression is transforming the signal zone of image and the main energy of image transformation mostly concentrate on few transformation coefficient, the image compression will be realized by quantification coding this part of transformation coefficient。 

2。The image compression under DCT: DCT discrete cosine transform is a kind of formation which can turn image into even function with image boundary fold by using Fourier Transform。 As a two-dimension Fourier Transform, it can keep cosine term and then compress image by removing the redundant space from image。

Keywords:  Image segmentation; MATLAB; The wavelet; DCT

目  录

第一章 绪论 1

1。1 图像压缩的背景 1

1。2 压缩的目的和意义 1

1。3 图像压缩的研究现状