提出了一种红外弱小目标检测算法性能评估的方法。该方法通过假设的数学模型给出了评价指标的理论推导过程,得到了虚警概率与检测门限,检测概率与信噪比之间的关系,绘出了接收机工作特性(ROC)曲线。同时,给出了在目标与背景分布函数未知的情况下,利用红外图像序列绘出ROC曲线来评估算法的检测性能。利用该方法得到了四种典型的弱小目标检测算法的ROC曲线,比较了四种算法的检测性能。

A performance evaluation method for infrared dim target detection is proposed. The relationship between the false alarm probability and detection threshold is studied deeply through the hypothesis mathematical model, so as the relationship between detection probability and SNR, the detection probability and the false alarm probability. The ROC curves for performance evaluation in infrared dim and small target detection are also given, drawn by infrared image sequences when distribution function of target and background unknown. Simulation experiments evaluate the performances of the four algorithms using the ROC curves.