Research On Detection Technology Of Anvil Of The Cubic Synthetic Diamond Press Based On Machine Vision

The anvil is an important part of cubic hinge press for synthetic diamond.Because the anvil works in high temperature and high pressure environment for a long time,the anvil surface is prone to crack due to fatigue.If it is not found in time and continues to use,it will lead to anvil collapse accident,causing great economic losses,and even casualties.At present,most of the crack detection of the anvil is done manually.To overcome the shortcomings of the existing crack detection methods for anvil,this paper proposes a defect detection method for anvil based on machine vision.The main research contents are as follows:(1)Aiming at the problem of the narrow space of the press synthesis cavity,an anvil image acquisition device including a moving module,a rotating module,an image acquisition module and an image processing module was designed and built,and each module was designed in detail.To ensure that the acquisition device can complete the image acquisition of the anvil,the related knowledge based on kinematics and the DH method parameter modeling method are studied,and the kinematics model of the anvil image acquisition device is established,and the end motion of the image acquisition device is described.The relationship between the base coordinate system and the end of the acquisition device is obtained by using the homogeneous transformation matrix,and the inverse kinematics calculation is performed,and the mapping relationship between the joint space of the device and the Cartesian space is obtained,which is the subsequent anvil image the trajectory planning of the acquisition device has laid the theoretical foundation.(2)Aiming at the problem that the temperature in the synthesis cavity is still high after the pressure is relieved and the temperature is reduced,the image acquisition device is required to quickly acquire all the anvil images,the research gives an optimal path planning algorithm for acquisition based on the genetic algorithm.After numbering each anvil,link it with the TSP problem,and use the X,Y,Z axis to project the long side as the distance between the two anvils,and establish the distance matrix model between each anvil.Finally,the genetic algorithm is used to search and optimize the distance matrix to optimize the best movement route of collecting the anvil image,to achieve the purpose of fast drawing.(3)According to the motion stability and time requirements of the device during operation,the trajectory planning algorithm based on the genetic simulated annealing algorithm is studied.By annealing the mutation rate in the genetic algorithm,the local optimization ability of the genetic algorithm is greatly improved,and the genetic simulated annealing algorithm is used to solve the time-optimal solution of the 3-5-3piecewise polynomial interpolation trajectory planning.The results show that under the condition of satisfying the kinematic constraints of the device joints,the working time of the device is reduced by 30%,and the operating efficiency of the device is improved.(4)According to the crack characteristics of anvil,the algorithm of anvil crack recognition is studied.The weighted average method is used to gray the color image,the global threshold segmentation method is used to extract the target area,the canny edge extraction algorithm and morphological processing technology are used to smooth the edge contour,and finally the feature selection method is used to complete the crack detection.(5)The software system for crack detection of anvil is developed by using C #,Halcon,and other software,and the experimental verification is carried out.The results show that the system developed in this paper can meet the requirements of anvil crack detection.