| Coal industry is an important basic industry related to national economic lifeline and energy security.Belt conveyor is the key equipment of coal transportation,which plays an important role in coal transportation.Conveyor belt is the core part of belt conveyor.Because of the complex and bad underground environment of coal mine and its own strength problem,it is easy to have major accidents such as longitudinal tearing and belt breaking,which will not only affect the safety production efficiency,but also cause very heavy economic losses.Therefore,the primary problem to be solved in the research of belt conveyor safety protection is the detection of belt longitudinal tearing.If the early stage of belt longitudinal tearing can be detected and response can be made,the impact of belt longitudinal tearing on the mine production process can be greatly reduced.At the same time,starting from the root cause of the longitudinal tearing of the belt,if the overrun materials in the belt transportation system can be accurately identified and responded,it can greatly prevent the damage of the overrun materials to the belt and ensure the safe and stable operation of the belt transportation system.With the development of machine vision technology,in order to make up for this defect,the "202 belt conveyor intelligent material overrun identification and safety protection system”has been developed for a raw coal production workshop in Shanxi Province,which provides a scientific guarantee for the safety production,cost reduction and efficiency increase of the coal preparation plant.This paper is mainly based on the "202 belt conveyor intelligent material over limit identification and safety protection system",the main research work is as follows:1)Based on the vision of the belt conveyor intelligent materials overrun identification and safety protection system of the overall design,and the system software and hardware part of the selection and design.In order to overcome the great impact of environmental coal dust,a dust-proof protection device is developed.When the overrun materials exceeds the set value,longitudinal tear is detected,and someone is near the belt,the system can alarm and stop.2)The image preprocessing methods such as image graying,image filtering and image enhancement are studied.Through the analysis of the characteristics of the image of out of gauge materials,the effects of common image filtering and image enhancement are compared,and the appropriate filtering and enhancement methods are selected.The experimental results show that these image preprocessing methods are conducive to the retention of image feature information and improve the recognition of the image to be tested.3)In this paper,the object segmentation and detection of the image of the out of gauge materials of conveyor belt are studied.A segmentation algorithm of the image of the materials of conveyor belt based on the combination of gray threshold and edge detection and a method of calculating the pixel value of the materials image through the line gray distribution curve are proposed.The practicability and effectiveness of the method are verified by experiments.4)According to the research status of longitudinal tearing of conveyor belt in recent years,the longitudinal tearing detection method based on laser line light source is adopted,and the principle and algorithm of longitudinal tearing detection and location based on image feature extraction are studied.The laser line image enhancement method,corner detection algorithm,line detection algorithm and conveyor belt circle mark detection algorithm are tested through experiments Analysis and comparative study.5)According to the actual demand,the image processing platform system is designed.The overall framework,data table content and specific functional modules of the system are designed,and the key technologies used in the system design are studied.This paper has 77 figures,1 table and 101 references. |
PDF Full Text Request