基于局部复杂度和方差的自适应图像锐化算法

Adaptive image sharping algorithm based on local complexity and variance

为解决传统的锐化算法对噪声敏感,易在强边缘处产生伪影效应,且对灰度变化较小的微弱细节锐度不足的问题,提出了一种基于图像局部复杂度和方差的自适应图像锐化算法。考虑了图像局部灰度变化的频率和强度,针对图像灰度剧变区和级别丰富区,应用图像局部复杂度和方差自适应调节增益函数,在背景噪声与原图一致的情况下,有效提升了图像中弱边缘和纹理细节的表现力。仿真结果表明,该算法处理后的图像边缘突出,纹理清晰,较好抑制了伪影的产生和背景噪声的放大。

To address the problem that traditional sharpening algorithm was sensitive to noise and easy to suffer from artifact effects on strong edges,and its acutance was insufficient for weak details,a novel adaptive image sharpening algorithm based on local complexity and variance was proposed.The algorithm considered both the frequency and intensity of image gray value.For the regions in image where gray value was rich or changed intensely,local complexity and variance were adopted to dynamically control gain function,so as to effectively improve the weak edges or texture details of image when the overall background noise was consistent with the original image.Results of the simulation indicated that the enhanced image had clear edges and textures, and it could effectively restrain the artifacts and amplification of background noise.