Infrared and Low-light-level Visible Light Enhancement Image Fusion Method Based on Latent Low-rank Representation and Composite Filtering

Due to the problems of low brightness and contrast, lack of detail contour information and poor visibility in the traditional infrared and low-light visible image fusion algorithm, an infrared and low-light visible enhancement image fusion method based on potential low-rank representation and composite filtering is proposed. Firstly, the improved high-dynamic-range compression enhancement method is used to enhance the brightness of low-light visible images. Secondly, the infrared and enhanced low-light visible images are respectively decomposed by using a decomposition method based on latent low-rank representation and composite filtering, and the corresponding low-frequency and high-frequency layers are obtained. Then, the improved contrast-enhanced visual-saliency-map fusion method and improved weighted least squares optimization fusion method are used to fuse the obtained low-frequency and high-frequency layers respectively. Finally, the low-frequency and high-frequency fusion layers are linearly superposed to obtain the final fusion image. Compared with other methods, the experimental results show that the fused image obtained by the proposed method has abundant detail information, high clarity and good visibility.