铁路铁路路况识别系统的设计和开发-深圳大学.PDFVIP

铁路信息系统 铁路计 算 机 应 用 第15卷第8期 RAILWAY INFORMATION SYSTEM RAILWAYCOMPUTERAPPLICATION Vol.15 No.8 文章编号：1005-8451（2006）08-0022-03 铁路路况识别系统的设计和开发 崔建荣，阮双琛，关惠玲 （深圳大学 工程技术学院 ，深圳 518060） 摘 要：为了实现对多种复杂铁路路况的识别，需要采集路况图像并加以处理，以得到识别所需的图 像特征，最后通过模式识别辨别路况。由摄像机、采集卡、计算机及相关软件建立一套路况图像采集与识 别系统。探讨数字图像的采集、图像处理中同态滤波、对比度增强、优化二值化算法、膨胀操作、粒子滤 波等处理方法以及模式识别方法在铁路路况识别系统中的应用。经过现场验证，所设计系统识别准确率达 94.4%，识别效果理想，可以满足路况采集及识别的需要，是一种可以运用在铁路机车智能减振系统中的新 方法。 关键词：铁路路况识别系统；数字图像处理；铁路路况图像；模式识别；图像采集 中图分类号：TP39 文献标识码：A Design and development of Automatic Railway Condition Recognition System CUI Jian-rong , RUAN Shuang-chen, GUAN Hui-ling (School of Engineering and Technology of Shenzhen University , Shenzhen 518060 ,China) Abstract: In order to get the feathers and recognize some complex statu s of railway, digital image should be gathered and processed. The railway image was taken by a CCD camera, and stored in the computer. Those images were processed so it could be get the feathers. The processing included homomorphism enhancement, contrast ratio enlargement, threshold, dilate, particle filter. The feathers of the image were used to recognize the status of railway. It was processed a series railway image, and the accuracy of recognizing was 94 .4%. The System was credible. It was a new method which could be used in Shocking Ab sorption System on engine. Key words: Automatic Railway Condition Recognition System; digital image processing; railway image; pattern recognition; digital image gathe