Infrared Weak Dot Target Tracking Algorithm Based on Meta-Learning

This article focuses on the problem of infrared point recognition in infrared weak point target tracking that traditional methods cannot solve, and the tracking results are unstable. Therefore, a meta learning algorithm is proposed for infrared point recognition analysis of infrared weak point target tracking. Firstly, theoretical analysis is conducted using binary datasets, and classification is carried out according to the requirements of infrared point recognition to reduce the problem of target loss caused by occlusion in infrared point recognition. Then, infrared point recognition is performed on the infrared weak point target tracking data node, and the infrared point recognition data is annotated to comprehensively analyze the recognition results of the infrared points. the meta learning algorithm outperforms traditional methods in terms of infrared small point recognition accuracy and infrared.