基于VC的图像压缩编码技术的研究及算法实现.doc

摘要 长期以来，人类获取到的信息其80％来源于图像媒体，15％来源于语音。所以，人们在自然界感受到的最主要的信息是视觉信息，这说明图像是人们生活中信息交流最为重要的载体，也是蕴含信息量最大的媒体。由于未经处理的图像信息量非常大，大力研究和开发图像压缩编码技术就非常重要。 图像之所以可以被压缩，其根本原因就在于图像数据具有较多的信息冗余。各个像素点之间不论在行的方向还是在列的方向上都具有很大的相关性，因而整个图像数据就具有相当大的冗余度，也就有相当的压缩潜力。再者，图像品质的评价取决于人眼，而人眼由于其视觉特性对图像的细微差别还是很难分辨的。因此图像数据还具有一定的视觉冗余，也就是说，在对图像数据进行压缩时，可以允许一定程度的失真。 本论文主要是通过哈弗曼编码、香农-费诺编码和行程编码三种方法对同一幅图像进行压缩，对所得到的图像压缩比进行比较，最后得出每种压缩方法的优缺点。本论文首先介绍了图像压缩编码技术的概念、分类、特点、发展方向及应用等基础知识，介绍了开发工具VC++的特点，较详细的说明了对位图图像进行处理的基础知识，然后分别对三种编码方法的原理和算法进行了分析，最后通过整体程序的运行结果得出结论。 关键词：图像压缩；哈夫曼编码；香农-费诺编码；行程编码 Abstract For long term, 80% information the humanity gains originated from image media, and 15% from sound media. Therefore, main information through which people feel the nature is visual information. This illustrates that image is the most important communication carrier in lives and also is the biggest information media. Because image without processing obtains a great deal of information, it is extremely important to study and develop image compressing and coding technologies vigorously. The basic reason that image can be compressed is that image data have information redundancy. Each pixel point has a big relevance with others, thus entire image data have a quite great redundance and also have a great compression potential. Furthermore, evaluation of an image quality is decided by the human eyes. However, the human eyes have their visual characteristics, so it is very difficult to distinguish the image fine or not. Therefore the image also has certain visual redundancy. In other words, some distortion may be accepted in compressing an image data. My graduation project is mainly about Huffman Coding, Shannon - Fannon Coding, and Run Length Coding. I am applying these three methods to a same image, then giving advantages and disadvantages for each compression method by comparing compression ratio. This paper first introduces some elementary knowledge, such as concept, classification, characteristic, development direction and applica