Surface Defect Detection System For Galvanized Steel Strip Based On Image Recognition

With the continuous development of modern industrial technology,the production of high quality and high value-added industrial products is bound to be more and more demanding for the quality of cold rolled strip,which requires uniform chemical composition,uniform mechanical properties,small dimensional tolerance and good surface quality.At the same time,in recent years,the price of raw materials in steel has increased,overcapacity and various industries have been reorganized,resulting in more intense competition in the steel market.To some extent,the steel industry is looking for a breakthrough from iron and steel production technology,in order to reduce production costs and get room for development.The surface quality of cold-rolled galvanized strip is an important index for measuring galvanized strip steel,which has attracted much attention in iron and steel enterprises in recent years.The production technology of cold rolled galvanized strip is complex,and the factors affecting its surface quality are numerous.How to detect surface defects in time is the key.In this paper,based on the research and development of the surface quality control of the cold rolled strip in Anshan Iron and Steel Group,on the basis of the current situation and analysis of the theoretical research on the surface testing of the strip at home and abroad and the actual situation of the galvanizing field,a set of surface defect detection system for cold rolled galvanized strip based on image recognition is studied and realized.The main contents of this paper are as follows:(1)On the basis of the analysis of the characteristics of the cold rolled galvanized strip production process and the requirements of the defect detection system based on image recognition,the design of the hardware system(the selection of the core equipment and the site deployment scheme),the system communication design and the software system design is designed to form the overall realization of the system.Plan.(2)Research on image acquisition method of strip surface in the process of system implementation.The CCD camera(Charge Coupled Device Camara)is used to collect the strip image in the production process continuously.At the same time,the method of cutting edge,splicing and compressing storage of acquisition image is emphatically studied.(3)The preprocessing methods of image acquisition are studied,including image filtering,image feature extraction and image illumination compensation.At the same time,the image is identified by the preprocessed image,and the suspected surface defect is preliminarily determined.Then,the defect is confirmed by one by one,and the defect image is stored.The defect image is stored.The prediction is given to the production operator,and the base defect image library is updated after the confirmation.After the strip defect identification,the defect location information is obtained according to the track and synchronous signal in the PLC(Programmable Logic Controller)system of the first level production line automation system(Logic Controller).(4)Using VS2010 integrated development platform,with OpenCV(open source computer vision library)under the C++ programming environment,the strip surface image acquisition,image recognition and defect detection software are developed,and the final system is delivered to the field.The advantages of the system on line operation and the design and implementation of the surface defect detection system of galvanized strip are explained.