TBC: A Target-Background Contrast Metric for Low-Altitude Infrared and Visible Image Fusion
This addresses a critical issue for low-altitude UAV reconnaissance missions by providing a reliable real-time metric for image fusion, though it is incremental as it builds on existing contrast principles like Weber's Law.
The paper tackles the problem of traditional no-reference metrics failing in low-altitude infrared and visible image fusion due to a 'Noise Trap', where they incorrectly assign higher scores to degraded images; it proposes the Target-Background Contrast (TBC) metric, which demonstrates high semantic discriminability and computational efficiency on the DroneVehicle dataset.
Infrared and visible image fusion (IVIF) is a pivotal technology in low-altitude Unmanned Aerial Vehicle (UAV) reconnaissance missions, enabling robust target detection and tracking by integrating thermal saliency with environmental textures. However, traditional no-reference metrics (Statistics-based metrics and Gradient-based metrics) fail in complex low-light environments, termed the ``Noise Trap''. This paper mathematically prove that these metrics are positively correlated with high-frequency sensor noise, paradoxically assigning higher scores to degraded images and misguiding algorithm optimization. To address this, we propose the Target-Background Contrast (TBC) metric. Inspired by Weber's Law, TBC focuses on the relative contrast of salient targets rather than global statistics. Unlike traditional metrics, TBC penalizes background noise and rewards target visibility. Extensive experiments on the DroneVehicle dataset demonstrate the superiority of TBC. Results show that TBC exhibits high ``Semantic Discriminability'' in distinguishing thermal targets from background clutter. Furthermore, TBC achieves remarkable computational efficiency, making it a reliable and real-time standard for intelligent UAV systems.