关键词: 双色红外, 目标识别, 分类器融合, 信息融合, 决策级融合

Abstract: Aiming at the problem of ATR for the dual band IR imaging system，a method of target recognitionis proposed using multi-features and multi-classifiers decision level fusion.The algorithm firstly extracts shape features and appearance-based features；Following，multiple classifiers are devised based on the different features to classify the target；Then，the target classification results of multiple classifiers are fused using proposed strategies of multi-classifiers decision level fusion；Finally，the multi-classifiers fusion classification results are deal with using devised decision rules to get the system’s target recognition results.The method makes good use of the multiple kinds of classification feature information of target in multi-sensors images，improves the performance and accuracy of target recognition to some degree.Experimentation results demonstrate the effectiveness of the proposed algorithm.

Key words: dual band IR, target recognition, classifier fusion, information fusion, decision level fusion