Fusion of infrared and visible images: A novel approach using fuzzy PID control and least-squares detail enhancement

The aim of fusing infrared and visible imagery is to enhance the informational content of individual images, enhancing their comprehensiveness and usefulness. The fused images of existing algorithms are easily biased towards one of the source images after fusion with a specific ratio. To overcome this challenge, this study proposes a novel fusion method based on fuzzy PID control system and least squares optimization(WLS) for detail enhancement. Firstly, the contrast information of the source image is measured, and the contrast difference is divided into two categories, and the two types of images choose different fusion strategies. The measurement results adaptively adjust the output of the fuzzy PID control system through feedback, so as to adjust the fusion weights of the classification and source images, and obtain the basic image after stabilization. In addition, four filters are used to extract details of source images, and the details are enhanced by WLS optimization. The final fused image obtained by combining the two can capture infrared and visible information to a greater extent. Experiments show that our method has good visual effects in dealing with boundary artifacts, and has significant contrast and rich texture.