The Detection Method Based On Mid-infrared And Long Infrared Vision For Conveyor Belts

There are important safety problems of coal transportation in coal mining.Detecting out the safety hidden danger in coal transportation is the key to ensure the safety of coal transportation.Belt conveyor has been widely used in underground coal mining due to its advantages of high reliability and strong practicability.The longitudinal tearing of the belt of the conveyor is one of the most important problems in coal mine transportation safety.Belt conveyor is an important link to maintain the production of fully mechanized coal mining face and output in the groundConveyor belt is an important part of the whole belt conveyor.Steel wire rope core conveyor belt is usually used in large capacity,high-speed,long-distance coal transport.In the process of coal mine transportation,hard impurities such as flake gangue and metal bolt may scratch or even tear the conveyor belt and cause longitudinal tear.Generally speaking,the damage distance of the conveyor belt with longitudinal tear is relatively long,and the whole conveyor belt needs to be replaced during maintenance,which will cause a long time interruption of production.Therefore,the longitudinal tear detection of conveyor belt is a key problem that must be solved in coal mine productionIn this paper,based on machine vision and combining the complementary characteristics of medium infrared vision and long infrared vision,a conveyor belt longitudinal tear detection method based on medium infrared vision and long infrared vision is proposed to improve the accuracy of visual detection method in the underground working environment.The dual-band infrared vision sensor designed in this paper is used to obtain medium infrared images and long infrared images of the lower surface of the conveyor belt.Tear,scratch and tearing precursor characteristics can be obtained from the collected image analysis.This paper studies the imaging principle of dual-band infrared vision sensor and divides the design thinking of dual-band infrared sensor into two parts:image signal acquisition module and image signal imaging module.The signal acquisition section is designed with a unique optical path and custom beam splitting films.In the signal imaging part,an infrared photoelectric sensor combining CCD and infrared focal plane array is designed.In C++compiling environment,the imaging process is simulated by OpenCV library to verify its feasibility and effectiveness.In the end,according to the experiments this paper verified that the proposed detection method has a higher accuracy in dark environment than the previous IBVD vision detection method.