摘要
由于雾霾天气的影响,对单幅监控图像存在着清晰度不足的问题,针对同态滤波和小波分析的优点,采用小波域变换对含有雾霾信息的图像进行多尺度分解,利用改进后的高通滤波器对小波的高频系数进行增强处理,低频系数部分先采取同态滤波的方法去除低频域的噪音,然后利用线性分段函数来拉伸低频信息,从而达到去除雾霾,提高图像对比度的目的。图像处理后结果表明,与单一的同态滤波和小波分析相比,改进的算法能有效地提高图像的清晰度。
Due to the influence of haze weather, there is a problem of lack of clarity for single image monitoring. Based on the advantages of homomor- phic filtering and wavelet analysis, carries out the multi-scale decomposition of haze images by wavelet domain transform. The high fre- quency coefficients of the wavelet are processed by the improved high-pass filter. The low frequency coefficient part adopts the method of homomorphic filtering to remove the noise in the low frequency domain, and then uses the linear segmentation function to stretch the low frequency information, so as to achieve the purpose of removing the haze and improving the image contrast. Experimental results show that compared with a single homomorphic filter and a single wavelet analysis, the improved algorithm can effectively improve the image clarity.
基金
西安市碑林区科技局项目资助(No.GX1413)
关键词
图像增强
同态滤波
小波分析
雾霾
Image Processing
Homomorphic Fihering
Wavelet Analysis
Haze
