Target detection method under unstable visible light-infrared image quality

The invention relates to a target detection method under unstable visible light-infrared image quality. The method comprises the following steps: firstly, reconstructing unstable target information of visible light and infrared images to obtain an incremental information image; fusing the incremental information image with the original visible light and infrared images to obtain visible light and infrared images containing incremental information; inputting the visible light image and the infrared image into a feature extraction network, and extracting global features of the visible light image and the infrared image; inputting the global features into a kernel similarity measurement module to obtain a similarity weight of a balance optimization network; multiplying the similarity weight by the global feature to obtain a balance optimized feature; inputting the features into a feature fusion module to obtain a fused feature map; and inputting the feature map into a detector for target detection. According to