Research And Implementation Of Intelligent Monitoring System For Inland Vessels Based On Computer Vision Technology

The development of inland waterway transport in South of Jiangsu is very prosperous.High quality,intelligent monitoring and management of inland waterway vessels has become an important issue in the field of water transport management.In order to effectively improve the water security management and anti control ability,public security departments actively carry out water intelligent monitoring system construction,through a wide range of sensing technology,modern communication technology,intelligent video technology and on inland waters of the main channel,the key areas of real-time monitoring control,via the record of the ship's automatic capture,in order to achieve "reflect the real-time dynamically and quickly find the suspicious targets,providing verification crime clues and evidence" etc.three major business functions.CCTV system,which is different from other administrative departments,is concerned with the purpose of traffic flow and channel dynamics.The intelligent monitoring system established by the public security department requires the recording and analysis of the specific details of the monitoring target.Research papers and put forward water intelligent monitoring system installation methods,and with existing hardware and software resources based.Research and implementation of the four kinds of intelligent video image processing algorithm,proved suitable for effective ways and means of inland river ship real-time monitoring and intelligent analysis.The main research work of this paper includes:(1)studying the Inland Ship HD picture capture method,detailed discussion of intelligent monitoring erect geographical environment,hardware configuration requirements and intelligent capture process,as well as by the front-end and back-end two parts water composition of intelligent video surveillance system architecture.(2)study the water moving target detection method:frame difference algorithm,Gaussian mixture algorithm,codebook algorithm and vibe algorithm four common background modeling algorithm,describes the four algorithm principle and the detection process,a brief discussion of the advantages and disadvantages of each,and the experiment than the adaptation degree of four algorithms of surface target detection.(3)discuss the motion of ship target tracking,using morphological method to improve binary foreground image accuracy,with seed filling method in a labelled graph blob targets,by using Kalman filter to target tracking and matching,approximately trajectory with virtual channel to normalize the ship motion,collision detection in virtual line to command the range camera to capture the high quality ship target image,the capture process.(4)summed up the inland ships,several main characteristics,selected the hull size of the ship and the nameplate text two for algorithm of intelligent video processing features,through measurement of video and image feature analysis means to achieve the automatic extraction and records of these two characteristics.(5)the software and hardware environment of the program is set up,and the intelligent video analysis algorithm is realized by using the open source video processing libraryThe effectiveness of the intelligent analysis method is verified by the actual operation test and the movement result comparison.