论文部分内容阅读
利用红外目标同时具有位置、灰度、面积等多特征的特点,提出了一种基于多特征融合的目标关联算法。首先在极坐标系下对目标位置采用概率数据关联算法计算候选目标的关联概率,然后结合目标的灰度、面积特征的预测误差计算关联波门中的候选目标在各种特征条件下的关联概率,进而利用多特征融合方式,计算出综合关联概率,完成目标状态估计的更新。实验仿真结果表明,由于跟踪关联概率由多种特征共同确定,避免了目标位置特征信息不稳定所造成的跟踪精度下降的问题,实现了密集杂波环境下红外目标稳定跟踪,其跟踪精度和稳定性明显高于依靠位置特征信息进行关联的传统概率数据关联算法。
Based on the characteristics of infrared targets, such as position, gray scale and area, a target association algorithm based on multi-feature fusion is proposed. Firstly, the association probability of the candidate target is calculated by using the probabilistic data association algorithm in the polar coordinate system. Then, the association probability of the candidate target in the correlated target is calculated by combining the target gray-scale and the prediction error of the area feature , And then use the multi-feature fusion method to calculate the overall correlation probability and complete the update of the target state estimation. The experimental results show that the tracking probability is reduced due to the instability of the target location feature information because the correlation probability of tracking is determined by a variety of features, and the tracking of the infrared target is achieved with dense clutter. The tracking accuracy and stability This is significantly higher than the traditional probabilistic data association algorithm that relies on positional feature information for correlation.