【摘要】 提出了基于Bandelet变换域的图像自适应增强的新算法,利用Bandelet变换在表示二维图像线、面奇异性时的优越性,实现了抑制噪声和凸显细节间的有效均衡。按照最小化逼近误差原则,寻找出二进剖分块的几何方向,在最小化均分误差(MMSE)的原则下合并二进剖分块,寻找出图像增强的方向。根据图像增强方向将Bandelet块分为两类:有几何方向的Bandelet块和无几何方向的Bandelet块,并分析了这两类Bandelet块系数的不同特征,从而区分出噪声和信号、清晰边缘和脆弱边缘。在此基础上提出了一种新的增强函数,在抑制噪声的同时,增强较弱细节并保护图像中的清晰边缘不失真。实验结果表明,与传统的图像增强算法相比,该算法在抑制噪声和放大细节特征两方面均有明显改进。更多还原

【Abstract】 A novel algorithm based on the Bandelet transform field for adaptive image enhancement was proposed.The algorithm reaches effective balance between noise limiting and detail enhancing with Bandelet transform’s superiority in presenting the linear singularity and the surface singularity of planar image.First,the geometrical direction in each dyadic square was found according to the principle of the minimal approximation error,then the small dyadic squares were combined,which was based on the principle of Minimum Mean Square Error(MMSE),finally,the directions of image enhancement were determined.And according to the directions,the Bandelet blocks were classified into two sorts:the Bandelet blocks with geometrical directions and the Bandelet blocks without geometrical directions,and the different characteristics of the two sorts of Bandelet blocks were analyzed in order to distinguish the noise and signal,and the weak contrast features and the strong contrast features.Image enhancement was realized by a new strategy that can emphasize features with low contrast and protect the strong contrast features from distortions while restraining noise.Results of experiments show that this method offers significant improved performance in restraining noise and emphasizing features with low contrast over the conventional image enhancement algorithms.更多还原