基于自组织特征映射的图像压缩技术

计算机人工神经网络技术提供了新的图像压缩方法。自组织特征映射人工神经网络就能够用于图像的有损压缩。通过将图像分成若干小块,然后使用神经网络进行训练达到特征向量自动聚类,从而将这若干个图像块分成不同的类,其类别个数远小于图像块的个数,最后使用一个映射表保存这些信息。该方式,将图像中相同或者非常相似的部分归为一类,降低了信息冗余度,从而可以进行图像的有损压缩。该方法采用了计算机神经网络,有比较好的适应性,能够方便的和其它压缩技术结合实现效果更好的混合压缩,具有良好的推广价值。

Self-organizing feature maps based image compression technology

Abstract Self-Organizing Feature Maps(SOFM) artificial neural network can be used in image compressing process.Key technologies include unsupervised learning vector quantification,clustering,and vector matching.Neural network is used in classify image blocks which are split from the source image.The number of classes is much less than the number of image blocks.The final purpose is to produce an acceptable codebook for image compression.After the process,the information redundancy rate of source image will be reduced and a comparatively small destination image can be generated.SOFM is proved to be an applicable method for image loss compression.Neural network provides the feature of adaptability for the image compression process.Further more,the above-mentioned method,combining with other image compression technologies,is a convenient way to improve compression rate and quality.