The Measurement Method And System Of Mining Pick Wear Based On Shape From Focus

Mining pick is a part that directly contacts with coal and rock in coal mining machinery.Its performance and service life directly affect the efficiency of coal mining and the service life of coal mining machinery.The mining pick will suffer the overload,torque and stress produced by the interaction with coal.Its failure forms include the pick surface wear,fracture,deformation,and so on.The wear is the main form of the failure of the pick.The study of the wear degree of the mining pick can remind the staff to replace the pick in time and provide new technology for the production of the mining pick.Due to the limitation of measuring equipment,the traditional method of measuring the wear of mining pick is cumbersome and has low accuracy,so it needs higher requirements for measuring the wear area of mining pick.Shape from focus technology can compensate for the shortcomings of traditional wear measurement methods.It has the advantages of no occlusion,strong practicability,high accuracy and simple algorithm.A new method is proposed for pick wear measurement.Aiming at the problem of low restoration accuracy of traditional shape from focus algorithm,normal distribution operator filtering image processing method and improved focus evaluation function are proposed.Then,by integrating the existing algorithms and optimization methods of shape from focus technology,a mining pick wear measurement system is developed and accuracy of system measurement results is verified by reverse engineering technology.A normal distribution operator filtering image processing method is proposed based on median filtering.This method can detect the noise points in the image through the mean and variance of the normal distribution,screen the non-noise points in the image,filter the noise points in the image by median filter,and preserve the original information of the image to the greatest extent.It is advantageous to the follow-up processing of the image.In addition,based on the gray difference function,an improved focusing evaluation function is proposed,which takes into account the influence of all the pixels in the neighborhood on the central pixels.The normal distribution operator image filtering method and the improved focus evaluation function are applied to the study of mining pick wear measurement,which can accurately restore the three-dimensional surface topography of mining pick.The results show that the focus evaluation algorithm proposed in this paper has better accuracy,feasibility and effectiveness,compared with the traditional three-dimensional morphology restoration algorithm.The connection of different modules is realized and a mining pick wear measurement system based on focused shape from focus is developed through integrating the existing algorithms and optimization methods of different key technologies of shape from focus restoration,choosing appropriate programming language to design the system framework and system functions,constructing different functional modules.The hardware equipment is used to build an image acquisition test bench to collect sequence images of mining pick.Then the images are imported into the system to reconstruct the three-dimensional shape of the mining pick through setting parameters,the three-dimensional surface shape recovery map of the mining pick is obtained,which can realize the measurement of the volume of the mining pick and provide an accurate and efficient measurement platform for the measurement of the pick wear.The three-dimensional solid model of the conical part of the top of the mining pick is reconstructed by reverse engineering technology.The center of gravity of the model is merged with that of the original three-dimensional entity model and Boolean operation is performed to obtain the common entity of the overlapped part.The volume difference between the reconstructed three-dimensional solid model and the original three-dimensional model and the public entity is calculated separately,and the sum of the two volumetric differences is obtained.The recovery accuracy of the system is obtained by comparing with the volume of the original three-dimensional model.The experimental results show that the accuracy of the system measurement results obtained in this paper is over 97.5%.