论文部分内容阅读
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 method to evaluate the performance of infrared weak target detection algorithm is proposed. The method gives the theoretical derivation of the evaluation index through a hypothetical mathematical model, and obtains the relationship between the false alarm probability and the detection threshold, the detection probability and the signal-to-noise ratio, and plots the receiver operating characteristic (ROC) curve. At the same time, given the target and background distribution function is unknown, the use of infrared image sequences drawn ROC curve to evaluate the detection performance of the algorithm. Using this method, ROC curves of four typical weak small target detection algorithms are obtained, and the detection performance of the four algorithms is compared.