基于无线与图像的家用仪表数值识读装置*

作者：赵立蒙，张云洲，白秋石，师恩义，齐子新 日期：2009-11-16/span> 浏览：3105 查看PDF文档

基于无线与图像的家用仪表数值识读装置*

赵立蒙，张云洲，白秋石，师恩义，齐子新
（东北大学 信息科学与工程学院，辽宁 沈阳 110004）

摘要：针对当前人工抄表方式存在的入户难、不安全等问题，结合数字图像处理和无线数据传输，设计了家用数字型仪表的数值识读装置。该系统采用CMOS图像传感器读取家用仪表的数值图像信息，由高性能数字信号处理器（DSP）进行数字图像处理，采用模糊神经网络算法对仪表数值字符进行识别；识别结果通过近距离无线ZigBee通讯发送至集中器，以进行后期的收费、监控等服务。实验结果表明，系统对典型家用仪表数值的识别正确率高达95%，对一幅含6个数字的图像的识别时间大约为2 s，可以满足一般应用；此外，该系统还具有良好的可靠性和抗干扰性。
关键词：家用仪表；图像识别；数字信号处理器；神经网络；ZigBee
中图分类号：TP277文献标识码：A文章编号：1001-4551(2009)10-0054-04

Digital meter identifier based on wireless communication and images
ZHAO Limeng, ZHANG Yunzhou, BAI Qiushi, SHI Enyi, QI Zixin
(College of Information Science and Engineering, Northeastern University, Shenyang 110004, China)
Abstract: Aiming at the inconvenience and insecurity brought by current prevailing manual meter reading, this digital meter identifier based on wireless communication and image was designed. CMOS image sensor was used to acquire the image of domestic meters value and highperformance digital signal processor(DSP)  was applied to figure processing and identification. The result were sent to concentrator through the shortrange wireless ZigBee communications, so as to collect the fees, monitoring etc.. The experiments show that the rate of identifying correct numbers can reach 95%. An image with 6 figures will be identified within 2 seconds, which is enough for common use. Besides, the system takes the advantages of good reliability and antijamming.
Key words: household meter; image identification; digital signal processor(DSP); neural network; ZigBee
参考文献(References):
［1］刘百德.智能水表应用过程中存在的问题与解决方案［J］.中国住宅设施，2007(11):46-48.
［2］谭思云，罗俊生.远传水表抄表系统的设计［J］.中国水运：理论版，2007，5（1）:115-117.
［3］CHANDLER T. The technology development of automatic metering and monitoring systems［C］//The 7th International Power Engineering Conference,2005:144-147.
［4］唐轶峻，申小阳，朱雯兰.基于BP神经网络的数显仪表数字字符识别系统［J］.电测与仪表，2005，42（9）：42-45.
［5］王权平，王莉.ZigBee技术及其应用［J］.现代电信科技，2004(1):33-37.
［6］GISLASON D, GILLMAN T. ZigBee wireless sensor networks［J］. Software Tools for the Professional Programmer,2004(29):40-42.
［7］HECHTNIELSEN R. Theory of the Back Propagation Neural Network［C］//Proc. of IJCNN,1989(1):593-605.
［8］CHEN X Y. NNS ameliorative BP arithnetic realization on DSP［J］. Journal of Tianjin Technology University,2004,19(2):103-105.
［9］GAN W S, CHONG Y K, GONG W, et al. Rapid prototyping system for teaching realtime digital signal processing［J］. IEEE Transactions on Education,2000,43(1):19-24.
［10］HONG K H, GAN W S, CHONG Y K, et al. An integrated environment for rapid prototyping of DSP algorithms using matlab and texas instruments MS320C30［J］. Microprocessors and Microsystems,2000,24(7):349-363.