The Research On Computer Vision And Railway Cargo Gauge Monitoring Application

Machine Vision is a new and developing rapidly research domain of computer science and will has been applied in various fields, we can say "where need man vision, where need machine vision". The most important and absorbing is it is more useful in many occasions that man can not apperceive, for example, exact and quantificational sense, danger scene sense, and so on. The development of Machine Vision will impel the aptitude systems; broaden the research and application domain in which computer and all kinds of machine intelligence are used.The paper researches the early two parts of Machine Vision--low leveland middle level, which bases on Machine Vision, combines with image processing and image analysis. In the first part, it presents a threshold segmentation algorithm with image processing technology, to separate the target from the background and give the edge detection result of the target. In the second part, the main aim is to find the target's coordinate value in the World Coordinate System with some coordinate transforms. Two different methods arebrought forward--camera emendation and coordinate transform. The latermethod is verified by the railway cargo monitoring experiments, it is proved that there is dispersion between the calculated value and the actual value because of the experimental condition, it should be improved further.The research has been supported by Railway Ministry Transportation Bureau project " digital railway cargo monitoring system kernel algorithm and software study(2001-2003)"...