Вирішення проблеми засвітлення матриці тепловізора методами комп’ютерного зору

Abstract

This work is devoted to solving the problem of thermal imager lens overexposure in challenging conditions for computer vision tasks. The causes of image degradation are analyzed, and modern approaches based on image processing and machine learning are considered. The proposed methods improve the robustness and accuracy of computer vision systems operating under extreme conditions.