基于图像匹配和小波神经网络的RFID标签三维位置坐标测量法  

3D Spatial Structure Measurement of RFID Tags Based on Image Matching And Wavelet Neural Network

作　　者：俞晓磊 庄笑[2] 刘振鲁 YU Xiao-lei;ZHUANG Xiao;LIU Zhen-lu(National Quality Supervision and Testing Center for RFID Products（Jiangsu）,Nanjing 210029,China;College of Science,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)

机构地区：[1]国家射频识别产品质量监督检验中心(江苏),江苏南京210029 [2]南京航空航天大学理学院,江苏南京210016

出　　处：《测控技术》2018年第10期26-29,共4页Measurement & Control Technology

基　　金：国家自然科学基金资助项目(61771240);江苏省自然科学基金资助项目(BK20141032);中国博士后科学基金资助项目(2016T90452;2015M580422);国家质检总局科技计划资助项目(2017QK117;2013QK194)

摘　　要：针对现代智慧仓储物流中货物的出入库信息采集与货物盘点的需求,提出一种基于图像匹配和小波神经网络的RFID多标签网络三维分布结构坐标测量新方法。首先,利用垂直和水平双CCD系统从不同角度获取RFID多标签网络的图像;随后,利用小波阈值去噪的方法对图像中的噪声进行去除,利用模板匹配算法获取各标签的三维坐标;最后,基于小波神经网络建立了RFID多标签网络三维坐标分布与对应识读距离之间的非线性关系模型。利用实验获取的数据,对模型进行验证,实验结果表明,该模型平均预测相对误差小于0.0131,可为现场测试环境下实时预测多标签最优三维几何分布提供参考。In view of the demand for information collection and inventory checking of goods in modem intelli- gent warehousing and logistics, a three dimensional （3D） spatial structure measurement method for RFID muhi-tag network based on image matching and wavelet neural network（WNN） is proposed. Firstly, the vertical and horizontal dual CCD system was used to acquire the images of RFID muhi-tag network from different angles. Then the method of wavelet threshold de-noising was used to remove the noise in the image and the template matching algorithm was used to obtain the 3D coordinates of each tag. Finally, based on the WNN, a nonlinear relationship model between the 3D coordinate distribution of RFID muhi-tag network and the corresponding reading distance was established. The model was validated using experimentsal data. The experimental resuhs show that the average prediction error of the model is less than 0. 0131. This work will supply a reference for real-time prediction of optimal 3D structure of muhi-tag in the environment of practical testing scenario.

关 键 词：图像匹配 三维结构测量 RFID标签 小波神经网络 双CCD系统 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]