Video technology based secure and smart recognition method for railway crossing protection zone

Abstract

The invention discloses a video technology based secure and smart recognition method for a railway crossing protection zone. The method comprises the following steps: acquiring live video image data of the railway crossing protection zone in real time in a YUV mode; adopting a working codebook having characteristics of a Gaussian mixture model to perform motion detection on to-be-analyzed live video images to determine target objects; establishing an SVM (support vector machine) classifier and classifying the target objects in the to-be-analyzed live video images through the SVM classifier; judging whether the target objects are obstacles by using a multi-camera fusion coordinate transformation matrix. The video technology based secure and smart recognition method can perform secure and smart recognition on the railway crossing protection zone with more complicated geographical environment.

Description

technical field [0001] The invention relates to the technical field of safety devices at railway / highway crossings, in particular to a video technology-based safety intelligent identification method for railway crossing protection areas. Background technique [0002] With the development of domestic regional industrial clusters and regional urban agglomerations, the railway network used to connect these industrial clusters and urban agglomerations has also developed rapidly. my country has the longest and most densely distributed railway network in the world. At the same time, the speed and density of trains running on the railway network have been greatly increased. Railway crossing safety management is an important part of railway transportation safety, and the level of management directly restricts the smoothness and safety of the railway network. Railway crossing safety management includes three basic methods: manned crossing management, sound and light alarm managemen...

AI Technical Summary

Problems solved by technology

[0005] For example, in the Chinese patent document with the publication number CN101830218A, for the inspection of the brake, it is also used to monitor the brake lever of the truck under the static state of the truck. There are subjective factors, and it is difficult to control the technical problems of the monitoring results. A kind of Intelligent identification and monitoring system for train brakes, wherein the visual monitoring unit is used to continuously take pictures of the bottom of the carriage, and cooperates with the control unit to automatically identify and judge whether the brakes exist within the current visual range. The position in the picture, etc.; the swing unit is used to drive the swing of the visual monitoring unit, and the control unit controls and adjusts the position of the brake within the visual range of the visual monitoring unit to achieve the best shooting and measurement results; the visual monitoring of the program The unit is only suitable for brake inspection, not suitable for railway crossings with complex geographical environment

[0006] To sum up, the existing video-based intelligent recognition technology is only suitable for warehouse management, smart cars, production management, and parking lot management. Since it cannot recognize flat images such as shadows, leaves, and light spots, it is not suitable for geographical environment comparisons. Safety intelligent identification of complex railway crossing areas

Images