The Research On Railway Invasion Detection System Based On Intelligent Video Analysis

As a result of the imperfection of the railway infrastructure, there are many railway accidents at present. How to improve this situation has been very urgent. With the intelligent surveillance system as a modern security monitoring technology, in railway safety monitoring has been widely used, the video-based trespass monitoring system is studied widely and applied in railway safety protection as an attempt, which will reduce the number of the railway accidents.In this thesis, The author will analyze the railway intelligent video-based trespass system demand and functionality target, and system application structure will be put forwarded.The author have described the model of codebook which can be used to get the background of the railway scenes. In order to overcome the shortcoming of traditional frame difference method and statistical method of motion detection of the defective pixels, The author presented a method based on image block motion detection algorithm, and then carried out the visual invasion tracking models based on the Kalman filter.The railway intelligent video-based trespass system has been developed. This system can give simple semantic interpretation of railway invaders’behaviors.