Research On Velocity Measurement Method Of Belt Conveyor Based On Machine Vision

Belt conveyor is a kind of material transport equipment which is widely used in coal and industrial production logistics.As belt conveyor is one of the important equipment of industrial transportation,the actual running state can directly affect the overall industrial running efficiency.The running speed of the belt conveyor is the core parameter of its running state.It is of great significance to measure the speed of the belt conveyor under the premise of meeting the requirements of industrial field production.Under the existing industrial transportation environment,the belt conveyor speed measurement method is generally adopted the traditional contact velocimeter,which has high measurement accuracy and good real-time performance.However,due to the inherent physical properties of the contact velocimeter,there are some defects and deficiencies in the speed measurement process.First of all,the speed measuring instrument and the belt conveyor surface contact may have a skid problem.Secondly,due to the long-term contact of the contact surface,the loss of the speed measuring wheel will occur,which will affect the measurement accuracy.Finally,when the belt conveyor is transporting the material,the contact speed measuring instrument may collide with the material,thus affecting the production and transportation safety.Due to the above defects,in order to overcome the limitations of the traditional contact speed measurement method,this paper designs a non-contact belt conveyor speed measurement method based on machine vision based on machine vision theory.The theoretical basis of the velocity measurement method is as follows: when the belt conveyor is in the working state,its background image can be approximately regarded as a static image,and on this basis,the lower surface of the belt conveyor is marked with visual features.Through continuous shooting by high-definition industrial camera in unit time,displacement changes of visual features can be obtained.Speed changes of belt conveyor can be described by motion speed information of visual feature markers to a certain extent.The filtered and denoised belt conveyor image was obtained through image preprocessing,and the kernel function of Meanshift algorithm was applied to the YOLO detection target box to calculate the probability distribution of the characteristic values of the detected object.Then,the probability distribution was put into the calculated Meanshift vector to realize the detection and tracking of visual feature markers through iterative search.The transfer matrix is calculated by the camera calibration method,and the pixel displacement change of visual feature markers is mapped to the actual displacement change of three-dimensional space,so as to obtain the running speed of the belt conveyor.In this method,the optical camera system is used to collect the image of the conveyor belt and the visual algorithm is used for real-time analysis,so as to obtain the motion and displacement information of the belt conveyor and calculate the running speed.On the basis of the above theoretical research,the overall design of the method is carried out,including the design of machine vision algorithm and the design of the software and hardware of the experimental platform.Finally,the theoretical design method is verified by experiments.Experimental platform based on many experiments,the experimental results show that this design belt conveyor speed measuring method based on machine vision can overcome the defect of traditional contact velocimeter,at the same time,comparative experiments prove that belt conveyor speed measuring method based on machine vision precision and real-time performance meet the requirement of industrial production,and for belt conveyor running status detection innovation has certain significance.