一种新的遥感影像色彩均衡方法

A New Radiometric Algorithm for Color Balance of Remote Sensing Imagery

针对不同情况下拍摄的目标影像镶嵌产生辐射差异的问题,提出一种新方法消除这种差异。该方法首先在目标影像中选择特征点并进行灰度特征区域划分,然后对特征区域进行灰度信息统计,最后建立相应的亮度因子模型对整幅影像进行辐射均衡处理。同时,提出平滑型高斯滤波校正模型处理影像对接处的灰度偏差。实验结果表明,亮度因子辐射均衡处理影像后获得的标准差分别为0.1109和0.1261,均低于原始影像的标准差,降低了影像亮度反差,实现了影像的辐射均衡。该方法同样适用于灰度融合。

This paper present a new method to eliminate the radiometric difference after mosaicking of the target images in different conditions.The method firstly selects feature points in the target images to divide the gray feature regions,and then performs gray information statistics on each feature regions.Finally,it establishes the luminance factor model to perform radiometric balancing.The smooth Gaussian filtering correction model is used to deal with the image gray deviation.The experimental results show that the standard deviation obtained by the luminance factor radiation equalization method after processing two images are 0.1109 and 0.1261,respectively,which are lower than the standard deviation of the original image.The method can be applied to processing the images in different conditions and it can achieve the gray data fusion in various situations.