Development of Interference-Resistant Computer Vision Algorithms for Detecting Camouflaged Objects during Combat Operations

The use of computer vision algorithms for detecting camouflaged objects is becoming relevant in military operations because of the critical importance of accuracy and speed of recognition. The aim of the article is to determine the effectiveness of computer vision algorithms: CNN, YOLO, and ResNet in detecting camouflaged objects on the battlefield and to evaluate technological approaches for their optimization. The methodology included a comparative analysis of algorithms by accuracy parameters, image processing, and reliability in different lighting conditions, and scenario forecasting. The study identified key factors that affect camouflage detection, in particular, sensor quality, frame rate, and algorithm adaptability to environmental conditions. The results indicate the significant effectiveness of infrared and multispectral sensors in combination with high-speed algorithms, which increase the accuracy of object detection in dynamic combat conditions. The article outlines the practical importance of improving detection systems for military purposes, offering recommendations for integrating the latest technologies to ensure the reliability of operations. Further research may focus on improving the adaptability of algorithms to work effectively in extreme conditions.