摘要
近十几年来,分形(Fractal)在图像压缩技术中的应用已成为图像数据压缩领域中最为热点的问题之一。其压缩比在理论上可超过经典压缩方法的几个数量级。但实际上,原始图像经常被噪声污染。噪声的存在一方面使得图像编码的时间延长,另一方面降低了图像的信噪比,使图像质量明显下降。本文对小波变换、分形压缩编码的原理和特点进行了分析;结合小波变换和分形压缩编码,利用分形的自相似性,研究了基于小波域的分形图像去噪、压缩方法。该方法有效地减少了计算复杂度和编码时间并获得了良好的图像质量。实验结果表明,该方法在较大的压缩范围内,能够获得好的压缩结果,同时也表明采用这种方法的潜力之所在。
Over the last ten years, the application of fractal transform has become one of the hot problems in the realm of image compression techniques. Theoretically, the compression ratio of the fractal transform can be several orders of magnitude higher than that of the classic compression method. However, in fact, the original images can often be polluted by noise. The existence of noise can not only cause the image coding time to be prolonged, but also cause the signal-to-noise ratio of images to be reduced and hence cause the image quality to be decreased apparently. In this paper, the principle and characteristics of wavelet transform and fractal compression coding are analyzed. The method for fractal image noise reduction and compression based on wavelet are studied by using the fractal self-similarity. The method can effectively reduce the calculation complexity and coding time and obtain good image quality. The experimental results have shown that good compression results can be obtained by this method in the case of large compression ratio and the potential of the method is apparent.
出处
《红外》
CAS
2007年第5期26-30,48,共6页
Infrared
基金
中国科学院长春光学精密机械与物理研究所创新基金资助(ZJ99130B)
关键词
分形变换
小波变换
图像压缩
小波树
去噪
fractal
wavelet
image compression
wavelet tree
noise reduction
