一种基于局部方差改进的自适应USM算法

An improved adaptive USM algorithm based on local variance

针对智能家居监控系统中视频帧图像模糊的问题,提出一种与双边滤波、高斯滤波以及CLAHE算法强结合的非锐化掩膜算法。该算法利用双边滤波与高斯滤波,能够有效地提取视频图像中的高频部分,提升图像边缘细节表现力。对于图像灰度值剧烈变化的区域,根据局部方差构建自适应增益函数,动态调整图像各像素点的增强系数,对图像进行合理增益。实验结果表明,该算法比传统非锐化掩膜算法的峰值信噪比提升了28%以上,能够更有效率地锐化图像的边缘部分,保持较高的图像清晰度,且能够有效抑制图像中的低频段噪声,避免边缘过冲。

Aiming at the problem of video frame blur in smart home monitoring system,this paper proposes an unsharpening mask algorithm combining with bilateral filter,Gaussian filter and CLAHE algorithm.This algorithm can extract the high frequency part of video image effectively by using bilateral filter and Gaussian filter,and improve the image edge detail expression.For the area where the gray value of the image changes dramatically,the adaptive gain function is constructed according to the local variance,and the enhancement coefficient of each pixel of the image is dynamically adjusted to achieve the reasonable gain of the image.Experimental results show that compared with traditional unsharpened mask algorithm,the proposed algorithm can sharpen the edge of the image more efficiently,maintain higher image clarity,and effectively suppress the low frequency noise in the image to avoid edge overshoot by more than 28% in the peak signal to noise ratio.