Analysis of Quality Evaluation Indicators for Visible Light and Infrared Image Fusion Based on Target Detection

This paper introduces seven deep learning-based visible light and infrared image fusion methods and the specific definitions of traditional evaluation indicators. The performance of each fusion network is compared and analyzed from both qualitative and quantitative perspectives. Additionally, the paper evaluates the fusion networks from the perspective of object detection, aiming to provide some ideas and reference value for the selection of fusion methods.