For a comprehensive evaluation of the comprehensive performance of the adaptive camouflage system， a three-level model of set pair analysis is proposed， and the connection degree algorithm of indexes at each level and the grading method of comprehensive confidence are constructed. The two-level index system is refined from four aspects： technical indexes， service indexes， environmental indexes， and maintenance indexes， and the normalization method of each index is proposed. According to the real-time adaptability of the adaptive camouflage system to the background， the division criteria are put forward， and the grade intervals of each index are constructed. The weights of the two levels of indexes are obtained by the analytic hierarchy process （AHP）. The examples provided verify the feasibility and effectiveness of the method. The method can provide a reference for the application orientation of the adaptive camouflage system.