小波变换技术以其良好的空间－频率局部特性和与人眼视觉特性相符的变换机制，在图像编码领域得到了广泛的应用和研究。其中，针对小波系数的树形结构，已出现了许多经典的图像编码算法，但其还存在一定的不足，对其展开研究有着重要的意义。 本论文阐述了小波变换和图像编码的基础理论，分析了嵌入式小波零树EZW图像编码算法和其经典改进算法的原理；研究了图像小波变换中小波基的选择、分解层数和边界处理问题，在大量仿真结果基础上给出了一些重要结论；针对EZW算法的不足，提出了一种新的嵌入式图像编码方法—基于小波树能量的嵌入式图像编码算法。仿真结果表明，改进算法与EZW相比，重构图像的峰值信噪比(PSNR)在不同压缩比下都有一定的提高。 本论文的主要研究成果和结论有： (1) 对一般自然图像作小波变换时，采用（9，7）双正交小波并对图像作对称周期延拓能获得较好的重构质量。 (2) 提出了小波树能量的概念，并利用它来设计改进算法中小波树编码的扫描次序，实现了编码时的能量优先性原则。 (3) 针对EZW算法不能有效地消除图像小波变换域中的级内冗余，在改进算法中设计了一种新的扫描方法和有效值映射的表示方法。

The technique of wavelet transform has been applied and studied in the image coding field extensively because of its good feature of time-frequency and human visual system. Especially，aiming at the tree-structure of wavelet coefficients, many classical image coding algorithms appear in succession.But there are also lots of disadvantages for these methods, therefore the research on these methods is very significant. In the paper, firstly, the theories about image coding and wavelet transform are dissertated. And the embedded wavelet zero-tree image coding(EZW) and its improved algorithms are discussed and analysed in detail. Secondly, based on the research on the selection of wavelet basis、confirmation of decomposition stages and border processing，many important conclusions are concluded. Finally,a new image coding algorithm called“embedded image coding methods based on wavelet-tree energy”is offered and simulated by matlab emulation soft. The emulation results indicate the new algorithm can largerly improve the peak signal-to-noise ratio (PSNR) of the reconstructed image compared to EZW. The main research contributions and conclusions are summarized as follows. (1) obtain better reconstruct image choosing the bi-orthogonal (9,7) wavelet basis and symmetric periodic extension . (2) offer a conception “wavelet-tree energy” and use this conception to contrive scan-order of the improved image coding method. (3) contrive new methods about scan and virtual value mapping which can effectively eliminate the redundancy consisting in the same leave.