研究目的
To automate the monitoring of railway tracks for safe movement by detecting visible rail defects and foreign objects using stereo vision and blur analysis.
研究成果
The introduction of analysis of the image blur and the stereo vision allows to determine the characteristics of an object placed on the railway track during movement of the train, find out whether it is an obstacle or not, and to find the external defects of rail grid and determine its operational state. The method shows promise for integration into hardware-software control systems for railway safety.
研究不足
The range for a specific focal length is very limited, and there is a problem of 'duality' in determining the distance to the object. The method's applicability is constrained by the need for precise camera calibration and the physical setup on moving trains.
