Research On Intelligent Video Monitoring Method For Underground Belt Conveyor

With the rapid development of visual technology increasingly, intelligent videomonitoring technology has been applied to various fields, Currently the majority ofunderground video surveillance system monitoring equipment is carried out by manualmonitoring, has not yet reached the intelligent monitoring, the study of underground beltconveyor intelligent video monitoring technology has an important role.By analyzing the common faults of underground belt conveyor, we proposed speedmonitoring methods of underground conveyor belt and monitoring methods based on linearfeatures of underground conveyor belt, by monitoring the speed of the belt conveyor andconveyor line feature parameters to determine the functioning of the belt conveyor properlyand assisting manual monitoring.Image pre-processing method is suitable to study the underground video features. First,comparing a variety of image renderings de-noising filtering method by using noiseevaluation criteria, we use bilateral filtering method as an image de-noising method, and thenusing the multi-scale Retinex image enhancement algorithms to achieve mine belt conveyorvideo image enhancement. Experimental results show that the image filtering and imageenhancement effect is good, the enhanced image and a low luminance region of the highluminance region suppression.Speed monitoring methods underground belt conveyor is to determine whether thepresence of the target coal on the conveyor belt, if the target is based on the existence of coalbelt conveyor speed monitoring methods coal, except coal target is based on the existence ofthe proposed belt Conveyor speed monitoring methods corner. Running speed monitoringmethods of underground conveyor belt based on corners, the first use of the video imageframe difference method for background modeling to obtain an image of the foregroundimage, and then using the Shi-Tomasi corner detection algorithm detects corner from theforeground image feature, followed by the use of a pyramid LK optical flow method to track these corners, and finally calculate the upper corner point belt conveyor thus completing themovement speed of the conveyor belt status monitoring.After the underground movement of the conveyor belt deviation monitoring method isalso conducting image preprocessing, first using Canny operator edge detection, get the edgebinary image of the video image; secondly using edge detection conveyor line feature Houghtransform, in order to accelerate the image processing speed and reliability of the region ofinterest setting image; again for the belt edge breakage linear least square fit of the brokenline segments are combined into a complete straight line; final value of the slope of thestraight line and linear deviation in the X-axis intercept value monitoring belt conveyor.Experimental results show that the proposed method of intelligent video monitoring ofunderground belt conveyor can be used to monitor the state of motion of the conveyor belt.Intelligent video monitoring methods of underground conveyor belt can use fully undergroundvideo information, the realization of the belt conveyor smart automatic video surveillance.