采用模糊推理自适应加权融合的双色红外成像目标跟踪

Target Tracking in the Dual Band IR Imaging System Using Adaptive Weighting Fusion Based on Fuzzy Inference

针对双色红外成像制导系统中多传感器目标跟踪的实际问题，提出了一种基于模糊推理自适应加权融合的目标跟踪算法。该算法首先采用BP神经网络与模糊推理相结合的方法对各传感器的工作性能进行判决；然后根据各传感器的性能测度对多传感器测量数据进行自适应加权融合，得到目标状态的多传感器重建测量；最后采用卡尔曼滤波器对多传感器重建测量进行滤波得到目标状态的最终估计。实验结果证明了该算法的有效性和稳健性。

Aim at the problem of multi-sensors target tracking in the dual band IR imaging system, a method of target tracking is presented using adaptive weighting fusion based on fuzzy inference. The algorithm firstly decides the performance of all sensors using a method integrated the BP neural network and fuzzy inference; Then, sums multi-sensors observation data adaptively with different weights based on the measure of these sensors to get the multi-sensors reconstruction observation of target position; Finally, filters the multi-sensors reconstruction observation using the Kalman filter to get the system estimate of target position. The result of experiments proved the effectiveness and robustness of the algorithm.