Research On Visual Measurement System Of Refractory Brick Based On Line Structured Light

With the increasing penetration of information technology into industrial automation,many industrial product detection methods are gradually changed from manually and contact measurement detection to automated inspection,non-contact real-time measurement change.Machine vision detection,as a non-contact automatic detection technology,has been paid more and more attention in China.Based on the "visual measurement system for refractory bricks" in horizontal project,the key technology of refractory brick visual measurement system based on line structured light is studied in this paper.The visual measurement system for refractory bricks developed in this paper is mainly implemented in two aspects:the upper computer software module and the hardware platform module.The main content in this thesis include:Chapter 1;The existing inspection requirements of refractory bricks are analyzed.The development status of machine vision and 3D vision measurement technology is summarized.The current research status of machine vision in refractory brick detection is analyzed,and the research content of this paper is given.Chapter 2:The measurement principle of machine vision measurement system is described Combined with the requirements of this paper,the measurement principle and calibration of line structured light are studied,and the calibration model of vision sensor is given.The overall scheme of the fire brick visual measurement system based on line structure light is designed.Chapter 3:The image preprocessing of the refractory brick is completed,and Two methods of size measurement are proposed:Hough transform method based on K-Means clustering algorithm and polygon fitting algorithm.The precision and computation time of the algorithms are compared.Chapter 4:A method for identifying common defects of refractory bricks is proposed,A method of filtering in frequency domain is proposed for surface scratch defects.Based on the research of the image gray moment and height point cloud data,the tilt correction is completed.Finally,the depth defect recognition is completed by combining the segmented histogram of the height data and the connected domain extraction.Chapter 5:A brick measurement system based on line structured light vision is developed.From the system level,the functional modules of the upper computer software and the hardware platform of the developed measurement system are introduced.On the basis of this system,many repeated tests were carried out,and the data were analyzed to verify the accuracy and stability of the measurement system.Chapter 6:The main research contents of this paper are summarized,and the next research work of this paper is prospected.