Moving Object Detection, Tracking and Range Estimation in Infrared Videos using Deep Learning

Infrared video technology for moving Object Detection, Tracking and Range Estimation has become a pivotal tool in various fields such as video surveillance, infrared guidance, Unmanned Aerial Vehicle (UAV) based monitoring and autonomous vehicle systems to medical imaging and environmental monitoring. Detecting and estimating the range of a moving object in an infrared video is a critical task with applications in target tracking, obstacle avoidance, and 3D scene reconstruction. This abstract highlight the key aspects of R&D in the field of detecting and estimation of objects tracking and range detection using infrared video. The abstract sheds light on the emerging trends in object range detection, such as Computer Vision (Optical Flow) and Deep Learning for object detection, tracking which is useful for range estimation of a moving object in an infrared images time frames. The utilization of Neural Networks (NN) and Convolutional Neural Networks (CNNs) such as YOLOv8, Mask R-CNN and Faster R-CNN deep learning for Object Detection, LucasKanade and DenseNet Optical Flow techniques for object tracking and MonoDepth deep learning model for Range Estimation in infrared videos along with the proposed model is explored in detail.