A MODEL BASED ON ARTIFICIAL INTELLIGENCE FOR EVALUATING THE EFFECTIVENESS OF MULTIMODAL CAMOUFLAGE OF MILITARY OBJECTS AGAINST UNMANNED AERIAL VEHICLE SURVEILLANCE SYSTEMS

Abstract

This article proposes an AI-based model for evaluating the detectability of objects based on visible and thermal imagery captured by unmanned aerial vehicles (UAVs). The model integrates channel-specific assessments, contextual distinctiveness, and sensor quality metrics to calculate an integral risk factor and relative change. The numerical values provided serve as an illustrative example of the model’s operation and do not represent the results of practical experiments