基于分数阶微分和小波分解的图像增强
首发时间:2011-03-22
摘要:从分数阶微分和小波分解的特点出发,提出一种用于图像增强的方法,即首先使用小波分解方法分别多层次、多尺度分解图像,并重构出相应层次图像中的高低频成分,然后使用新提出的包含八个对称方向分数阶微分掩模算子有针对性地对分离出的高频、低频及原始图像信号分别进行处理,把处理的结果进行合并、叠加,同时深度地保留图像平滑区域的低频轮廓特征和非线性地保留灰度变化较大的高频边缘特征,对灰度变化不明显区域图像纹理细节也得到增强。从实验对比结果来看,此方法使用新算子相对于其他几种方法对增强图像平滑区域的复杂纹理细节效果要好。
For information in English, please click here
Image enhancement based on fractional differentials and wavelet decomposition
Abstract:Starting from the characteristic of fractional differentials and wavelet decomposition, a method of image enhancement was proposed, which firstly used wavelet decomposition method to decompose image,and reconstructed low-and-high frequency in correponding layer image, then it utilized the proposed new fractional differential mask operator on the eight symmeric directions to apart process the detached low-frequency, high-frequency and primal image signals. At the same time, it merged the results so that it could furthest preserve the low-frequency contour feature in those smooth areas, and nonlinely keep high-frequency marginal feature in those areas that gray-level changed greatly,and also enhance texture details in those areas that gray-level did not change evidently. Experimental comparison results show that the effect of enhancing complex texture details in smooth area by this method using new operator appears better than several other methods.
Keywords: fractional differentials wavelet decomposition image enhancement cover module
论文图表:
引用
No.4416517547953130****
同行评议
共计0人参与
勘误表
基于分数阶微分和小波分解的图像增强
评论
全部评论0/1000