论文部分内容阅读
针对红外弱小目标跟踪精度不高,实时性不好等问题,提出了一种基于灰色预测-概率数据关联滤波(GPPDAF)的弱小目标跟踪方法。该方法采用实时性较好的灰色模型对目标轨迹进行预测,并以预测位置为中心构建跟踪波门,然后通过改进的极大似然概率数据关联(MLPDA)算法计算波门内各量测点的概率权重,最后用各量测点的概率加权和作为跟踪结果。实验结果表明,该方法实时性能较好,较大地改进了红外弱小目标的跟踪精度。
Aiming at the problems such as the low precision of tracking infrared weak targets and the poor real-time performance, a weak target tracking method based on gray prediction-probability data association filtering (GPPDAF) is proposed. In this method, the real-time gray model is used to predict the target trajectory, and the tracking wave-gate is constructed based on the predicted position. Then the MLPDA algorithm is used to calculate the measurement points Of the probability of weight, and finally use the probability of each measurement point weighted sum as the tracking results. The experimental results show that the proposed method has good real-time performance and greatly improves the tracking accuracy of infrared weak targets.