Reserch On Conveyor Belt Tear Visual Inspction System Based On Otsu Algorithm

Because of the advantages of large capability,high speed and continuous transportation mode,belt conveyor is widely used in bulk and block material transportation system.Also,in the process of coal production and transportation,belt conveyor has lots of advantages which other transportation equipment has not.Therefore,the belt conveyor is one of important coal production and transportation equipment.In the long time and high load operation process,belt conveyor is faced with the risks caused by roller fault friction,metal object sticking,gangue scratch or other factors.Conveyor belt tearing often result in coal transportation system paralysis and cargo spill jam roadway accident.Serious Conveyor belt tearing cases will lead to casualties and adverse impact on coal production.In this paper,we study the applicability of several image processing algorithms in the detection system of the conveyor belt tearing based on deep understanding of machine vision system.Its purpose is to design an effective conveyor belt tearing visual inspection system and coal production to ensure the safe of transportation system.This paper mainly includes:1: At present,compared with other conveyor belt tearing detection methods,visual detection method has incomparable superiority.Research on the application of the existing machine vision system in the field of conveyor belt tearing detection,this paper presents a design scheme of the belt tearing visual inspection based on Otsu algorithm,which is practical,reliable,stable and real-time.2: Based on analyzing the theory of machine vision,structure,according to the design principles of coal production detection system,this paper put forward the conveying belt tearing visual detection system design,and deeply study the hardware structure,function of the system and all hardware which can realize the function of the system required for devices.3: The research of image processing algorithm is the core of the design of the conveyor belt tearing detection system.For the gray image captured by the industrial CCD camera,the following algorithm research will be three parts:Image preprocessing algorithm research;Image processing algorithm research;Tearing diagnosis algorithm research.The system uses the median filter and histogram equalization to preprocess the picture of the conveyor belt surface,which can remove the salt and pepper noise and enhance the image contrast.Image threshold segmentation is a key step in the process of image processing and a key step of system tearing detection.By study of the local threshold segmentation algorithm,iterative threshold segmentation algorithm,Otsu threshold segmentation algorithm,advantages and disadvantages of the three methods in image processing and image processing time,determine that the Otsu threshold segmentation algorithm of image processing is the best threshold segmentation algorithm as the conveyor belt tearing visual detection system.Finally,according to the characteristics of the conveyor belt surface two valued image,diagnostic rules and algorithms to achieve the purpose of system tearing detection.4: The software structure of the visual inspection system of conveyor belt tearing is proposed,and the system function is realized by software.Virtual instrument technology continues to develop,modular hardware structure design,simple graphical LabVIEW programming environment for the machine vision system has brought a new revolution.Based on the advantages above,the conveyor belt tear visual inspection system uses virtual instrument to realize system image processing,data communication,database storage and other functions.5: Relying on the existing belt conveyor equipment in the laboratory,for the system to achieve the function of selecting the appropriate industrial CCD camera,LED light source,speed sensor,the next machine PXI platform,PC and other hardware equipment to build experimental platform.In the belt conveyor 5m / s running speed,start the system,through the system interface to check the operating state of the conveyor belt surface image processing results,analysis,comparison system output normal tearing conveyor surface binary image data.Set the parameters of the tear diagnosis algorithm to prove the feasibility of the system.This system is the application of machine vision in the field of conveyor belt tear detection.The system obtains the surface image of the conveyor belt through CCD camera,and synthesizes the median filter,histogram equalization,Otsu gray threshold segmentation method to deal with it.System achieves the function of real-time nondestructive testing of the conveyor belt in the running state.