Research And Application Of On Line Casting Slab Surface Defect Detection Based On Laser Scanning Method

China's iron and steel industry is in the transition from large to strong.The requirements of carbon emission reduction and carbon neutralization,as well as the promotion of CC-DR production mode,all put forward higher requirements for the surface quality of continuous casting slab.The defects(internal defects and surface defects)produced in the production process of continuous casting slab can not be completely avoided.It is very important to monitor the surface quality of hot continuous casting slab in real time,record and classify the defects.The real-time monitoring of the surface quality of the hot continuous casting slab can avoid the meaningless processing of the defective slab,and improve the product yield of the steel plant.The real-time quantification of defects and the establishment of billet defect database are of great significance to the data management and analysis of the whole iron and steel process.In view of the above reasons,the on-line non-destructive testing procedure and defect data processing procedure of continuous casting slab become the necessary conditions for the normal operation of the hot delivery hot charging and rolling integration procedure.Aiming at the real-time detection of three-dimensional quantitative surface defects of high temperature continuous casting slab,a real-time detection method based on area array CCD and laser scanning is introduced,and field verification and application are carried out.In addition,machine vision and graphics processing methods are used to display the surface defect morphology of continuous casting slab.The specific research contents include: 1)by calculating the light intensity distribution of each band under high temperature slab,combining with the spectral response characteristics of area array CCD image sensor and the wavelength of laser light source,a calculation method of optimal imaging shutter time of high temperature slab surface laser line based on pixel gradient statistics is proposed,and the response time of area array CCD and the central filtering wavelength of filter are determined.2)Based on the technical parameters given by the steel plant,the parameter selection of main equipment is carried out.Combined with the size parameters of the laboratory samples,the size calibration of the experimental results is carried out,and the calibration results are applied in the actual field.In addition,the three-dimensional height map of the specimen surface under laboratory conditions is also displayed,which provides a basis for displaying the height map of the slab surface.3)Considering the environmental conditions of steel plant application,a linear fitting method based on weighted least square method is proposed.The field application results show that the method can effectively filter out the beating and tilting of the slab in the process of dynamic movement.4)Based on the real-time relative height image of defect slab generated by field operation,various algorithms such as gray-scale transformation,spatial domain filtering and frequency domain filtering are used to process the gray-scale image,which can remove the influence of image noise and enhance the 3D display effect of defects.