基于几种不变量融合信息的缺损目标识别

基于几种不变量融合信息的缺损目标识别 RECOGNITION OF OCCLUDED TARGET WITH FUSED DATA OF SEVERAL INVARIANCES 廖原 Liao Yuan 1 first-author 袁捷 Yuan Jie 1 赵恒卓 Zhao Hengzhuo 1 胡正仪 Hu Zhengyi 1 王延平 Wang Yanping 1 武汉大学电子信息学学院 武汉大学电子信息学学院 College of Electronic Information, Wuhan University College of Electronic Information, Wuhan University 论述了基于３种不变量的融合信息来识别缺损目标的方法．该方法采用Ｄｅｍｐｓｔｅｒ－Ｓｈａｆｅｒ证据推理方法作为决策层的融合工具，将拐点、线矩、高阶神经网络的分类结果进行信息融合．分类实验证明，该方法可以有效地提高系统的识别精度． This paper presents a method of recognizing occluded target with fused data,which comes from three kinds of invariances such as corner, moment and high order neural network. In this paper, Dempster Shafer Evidential reasoning is selected and used for data fusion at report level. The classification experiment shows that this method effectively improves the matching precision of the recognizing system. 缺损目标识别 不变量 信息融合 Dempster－Shafer证据推理 recognition of occluded target invariance data fusion Dempster Shafer evidential reasoning TN941.1 国防科工委预研基金 1998-01-01 2021-04-01 1