图像是多媒体应用中的重要媒体，而传输频带和存储空间的限制，使得图像压缩技术成为当前信息传输系统中最关键的技术之一。小波变换技术以其良好的空间-频率局部化特性和与人类视觉系统（HVS）相吻合的特性，在图像编码领域得到了日益广泛的研究和应用。 本论文主要研究了基于小波变换图像编码方法。首先概述图像编码基本理论和方法，介绍了一些常见的压缩方法和涉及图像、视频的国际压缩标准；其次介绍了小波变换及其应用于图像压缩的相关理论；针对静止图像的压缩编码，在嵌入式小波零树编码（EZW）方法的基础上，提出了一种具有分块预处理的改进算法。该方法首先将图像经过小波变换，产生不同分辨率下的各子带图像；然后对低频子带采用无损熵编码，在阈值 下对高频子带进行边缘检测，根据不同的情况对边缘、纹理和平滑区进行分类零树编码。与原始的嵌入式小波零树算法相比，这种改进的编码算法有利于更好地利用小波变换后的系数，有效的降低码率，同时增强图像压缩后恢复图像的主观视觉效果。实验结果表明，这种改进算法得到的重构图像的主观视觉效果良好，在相同码率的情况下，峰值信噪比（PSNR）较原始的嵌入式小波零树算法有较大的提高。本论文还将零树编码与分形编码算法相结合，对小波域分形编码算法进行了分析和研究。

Image is very important medium at application of multimedia. Because of limitation of transmission bandwidth and storage capacity, image compression became one of critical techniques at information transmission system. In recent decade, wavelet transform has been studied and applied extensively because of its good time-frequency features and tally with human visual system. In this paper, we focus on the wavelet transform code. Firstly, We summarize the basic theory on image code, including some familiar compression techniques and some international compression standards on image and video. Secondly, we introduce the basic theory on wavelet transform code. At last, a improved image compression algorithm is put forword about still images based on embedded wavelet zero-tree coding scheme. First of all, image is divided into some multi-resolution subband images by using wavelet transform. The low-frequency subband is coded with lossless entropy coding. The high-frequency subband are made edge detection. Different edge,grain and smooth area use different coding schemes. Compared with original embedded zero-tree wavelet algorithm, this improved coding algorithm can apply the wavelet transform coefficient effectively, low bit-ratio and intensify the visual effective of restruction imges synchronously. The results of experiment show that the reconstruction images are good, and the peak signal-to-noise ratio(PSNR) of the reconstruction images are improved greatly at the same bit-ratio. Furthermore this paper combined the EZW with fractal coding method, analysed the fractal coding algorithm based on wavelet transform.