A high-performance approach to detecting small targets in long-range low-quality infrared videos

Since targets are small in long-range infrared (IR) videos, it is challenging to accurately detect targets in those videos. In this paper, we propose a high-performance approach to detecting small targets in long-range and low-quality infrared videos. Our approach consists of a video resolution enhancement module, a proven small target detector based on local intensity and gradient (LIG), a connected component (CC) analysis module, and a track association module known as Simple Online and Real-time Tracking (SORT) to connect detections from multiple frames. Extensive experiments using actual mid-wave infrared (MWIR) videos in range between 3500 and 5000 m from a benchmark dataset clearly demonstrated the efficacy of the proposed approach. In the 5000 m case, the F1 score has been improved from 0.936 without SORT to 0.977 with SORT.