Design And Realization Of Breach Stay Detection System For Construction Vehicles In Oil Pipeline Area Based On Video Image

With the rapid development of urbanization,the interlaced pipeline resources play an indispensable role in the construction of the city.The pipeline is equivalent to the urban connective tissue,which is of great significance to urban development.Therefore,it is very important for the protection of the pipeline.Pipeline buried in the ground,the pipe above the ground often appear in the land site,often after a lot of vehicles,the weight of the vehicle under the ground will cause some pressure on the pipeline,compared to ordinary vehicles,the pressure on the underground pipelines engineering projects particularly serious,However,excavators again have the most serious impact on the underground pipelines.In recent years,there have been many major accidents.Excavators have caused great potential safety problems for underground pipelines.Therefore,an oil pipeline that can work automatically in real time is needed Regional engineering vehicle violation detection system to reduce manpower and resources to meet the requirements.Based on the technology of image visualization,the system combines machine learning with deep learning technology of convolutional neural network to monitor the video and the video acquired by the camera as input.Through the image analysis of the pipeline surface area and the use of cascade classifier method to find the rough area of the excavator to extract,the extracted area through the image median filtering method to eliminate noise,and uniform light processing,combined with Caffe trained model,the extraction of the regional image analysis,according to the set within the safety threshold time excavator stay in the pipeline to detect the situation to determine whether the excavator will cause potential safety problems on the pipeline,excavators on the illegal alarm issued a warning,notify the relevant personnel to deal with,take preventive measures,and timely warning of irregularities Excavator vehicles for transfer,to ensure the safety of the pipeline.Finally,the system has been tested for its function and performance.The test and analysis of the actual scene and the test data set show that the system has the advantages of prompt detection,fast response and high detection accuracy for pipeline excavators.On the irregularities of the excavator to determine the abnormal situation more accurate,promptly issued a warning.Thus verifying the feasibility and accuracy of the system,timely and effective.