Research On Tile Position Measurement System Based On Machine Vision

Chinese ceramic tile production and consumption account for half of the world market,and it has become the world’s largest ceramic tile base.Therefore,the ceramic tile laying industry has a very large market in China.The traditional tile laying method is to lay the tiles manually,which consumes a lot of manpower and time,and the quality of the laying is greatly affected by the proficiency and fatigue of the workers.With the development of robotics,tile laying robots have begun to replace manual tile laying.This paper proposes a tile position measurement system using machine vision technology for the tile laying robot developed in the laboratory to replace the original laser sensor tile position measurement system,and the angle and distance between the edges of different tiles are measured.This paper firstly calculates and selects suitable industrial cameras,lenses,communication methods and light sources according to the sight distance and accuracy requirements of the measurement part of the tile laying robot,and completes the hardware platform construction of the tile position measurement experimental system.Based on the original structure of the end of the robotic arm of the laboratory tile laying robot,an installation system is designed for the camera module according to the actual size requirements and strength requirements.In order to verify the strength and stability of the camera bracket under minimal mass,the force analysis and and modal analysis of the camera bracket is carried out.Secondly,in view of the problem that the brightness of the obtained image of the positional deviation between the tiles is too dark,the method of histogram equalization and the interception of the region of interest is adopted,the image contrast is increased,and the center picture of the edge of the tile is intercepted,and Then the edge of the tile is segmented from the background image using an adaptive thresholding method.Aiming at the problem that the Canny algorithm requires subjective judgment of the threshold limits,a Canny edge algorithm based on the Otsu algorithm is proposed.The Otsu algorithm is used to determine the upper and lower threshold limits,which reduces the error caused by manual determination of the threshold limits.Comparing the detection accuracy and speed of LSD for different input images,an LSD line detection algorithm based on edge detection is proposed.First,the edge of the tile is detected,then the linear expression of the edge of the tile is detected,and the distance and angle between the edges of the tile are finally measured.Finally,on the tile laying robot equipment developed in the laboratory,the tile position measurement system developed in this paper is experimentally verified,and the measurement results of the tile edge position under different algorithms are compared,which verifies the superiority of the algorithm in this paper.The measurement accuracy is better than the laser edge position.Experiments on the measurement system show that the distance measurement accuracy of this system is 0.2mm,and the angle measurement accuracy is 0.2°.