论文部分内容阅读
证据融合是提高目标识别准确性的有效方法.为了解决高度冲突证据融合时产生不合理结果的问题,提出了证据源权重的评定原则.引入了证据距离的概念,根据证据源权重的评定原则,提出了证据源权重的计算方法,实现了对各传感器证据的按权相加修正,从而在信息融合之前消除了冲突证据,避免了不合理结果的产生.为了有效利用先验知识,提高目标识别的效率,分两种情况设计了证据融合的方案,并进行了融合复杂性分析.分别对两种证据融合方案进行了仿真试验并进行了比较分析,仿真结果验证了使用证据融合进行目标识别的有效性.
Evidence fusion is an effective way to improve the accuracy of target recognition.In order to solve the problem of unreasonable results when highly conflicting evidence is fused, the concept of evidence source weight is put forward.The concept of evidence distance is introduced.According to the evaluation principle of evidence source weight, A method of calculating the weight of evidence source is proposed and the correction of the weight of each sensor evidence is added to eliminate the evidence of conflict before the information is fused and avoid the generation of irrational results.In order to make effective use of prior knowledge and improve the target recognition The scheme of evidence fusion is designed in two cases and the complexity of fusion is analyzed.The simulation experiments of two schemes of evidence fusion are carried out respectively and compared with each other.The simulation results verify that the target recognition Effectiveness.