Detection and classification of steel defects using machine vision and SVM classifier

Type : Article de conférence
Auteur(s) :  Rachid Zaghdoudi, Hamid Seridi, Adel BOUDIAF
Année :  2017
Domaine : Electronique
Conférence: 2nd International Conference on Automatic control, Telecommunication and Signals (ICATS'17)
Résumé en PDF :  (résumé en pdf)
Fulltext en PDF :  (.pdf)
Mots clés :  Defects steel, machine vision, pattern recognition, HOG, GLCM, SVM classifier

Résumé : 

the importance of quality control of steel products is increasing day by day in the manufacturing industrial systems because it offers the possibility of knowing the state of the products without stopping the production line which allows the control of a defect before it becomes a complex problem and avoiding production losses. Human quality control of steel products remains tedious, fatiguing, bit fast, bit robust, dangerous or impossible, therefore the use of automated vision system can significantly improve the quality inspection process, because the machine vision technology can overcomes the majority of manual inspection problems cited above and provide an interesting solution especially, with the impressive increasing of computing power of today's computers and the good quality of images that offer the current cameras.The main objective of this research is to propose an efficient control system based on machine vision technology and SVM classifier to classify different types of steel defects.