为了提高红外成像质量的同时更大程度地保持纹理信息,提出一种多尺度分析和加权最小二乘法的条纹噪声非均匀性校正算法.该算法利用加权最小二乘法对图像进行平滑, 应用小波变换提取平滑图像的垂直分量, 并将其垂直分量替换为原始图像的垂直分量, 利用小波重构输出校正后的图像.算法能够精准地去除红外噪声, 而不会带来更加麻烦的“鬼影”问题.用该算法对多组不同红外图像数据进行仿真实验, 并与其他先进的红外条纹非均匀校正算法进行对比分析, 结果表明所提算法校正结果有较好的视觉效果和图像质量评估参数.

In order to improve the quality of infrared images while maintaining texture information to a greater extent, a multiscale analysis and weighted least squares method for stripe noise nonuniformity correction is proposed. The algorithm uses the weighted least squares method to smooth the image, applies the wavelet transform to extract the vertical component of the smooth image which is replaced with the vertical component of the original image subsequently, and reconstructs the corrected image by wavelet reconstruction. The proposed algorithm can accurately remove infrared noise without causing more troublesome "ghosting" problems. The proposed algorithm is experimented on multiple sets of different infrared image data, and compared with the stateoftheart destriping algorithms. The results show that the proposed algorithm has better visual effects and image quality evaluation parameters.