An Infrared and Visible Image Fusion Method Based on Illumination Enhancement

A fusion method for infrared and visible light images based on light enhancement was proposed to address issues such as uneven light distribution, low brightness, insufficient contrast, and missing detailed information in low-light, visible-light, and infrared images. In the light enhancement stage, the infrared image is first processed using specific logarithmic and exponential functions to enhance the local contrast of low- and medium-intensity pixel regions. Second, the adaptive multiscale Retinex algorithm is adopted to enhance the overall brightness of low-illumination visible light images, improve detailed feature and texture information, and provide them with good observability. Subsequently, using the gradient transfer and total variation minimization methods, the two enhanced images were fused and reconstructed to obtain the final fused image. Experimental results demonstrate that the fused images produced by this method exhibit high contrast, uniform pixel distribution, clear detailed features, and good visual quality.