Improved cross pattern approach for steel surface defect recognition
- PDF / 1,141,078 Bytes
- 10 Pages / 595.276 x 790.866 pts Page_size
- 101 Downloads / 206 Views
ORIGINAL ARTICLE
Improved cross pattern approach for steel surface defect recognition Zoheir Mentouri 1
&
Hakim Doghmane 2 & Abdelkrim Moussaoui 3 & Hocine Bourouba 2
Received: 29 May 2020 / Accepted: 3 September 2020 / Published online: 19 September 2020 # Springer-Verlag London Ltd., part of Springer Nature 2020
Abstract In steel-making processes, different methods are used for online surface product monitoring. Such a control has become a necessity to avoid additional costs resulting from the poor quality of the final product. With the reported performance that varies from one application to another, all the applied methods have to meet a minimum of criteria as accuracy and speed. This effectiveness is assured thanks to a relevant image description and efficient defect classification algorithms. The Dual Cross Pattern technique, successfully applied in face recognition, is a concept that relies on coding pixels to provide such a discriminating description of the image. Its principle can perfectly be used in industrial vision applications for surface defect recognition. In this study, the relevance of this method of describing defect images is evaluated, and improvements are proposed to increase its efficiency. The experimental study shows that the pixel coding that considers the variations of the intensity in several directions and captures the information from more than one pixel-neighborhood level makes it possible to better detect the variability in the defect image and helps to increase the defect recognition rate. The experiments are carried out with the use of the published Northeastern University (NEU) database for the comparison and with a new constructed database to better show the improvements brought by the proposed approach. Keywords Computer vision . Image description . Surface defect classification . Steel process
1 Introduction In recent decades, vision machine–based systems have been the tool widely adopted for the inspection of steel products. Indeed, to detect and recognize steel surface flaws, many approaches have been implemented with an effectiveness varying from one application to another. The key step in such an application is the extraction of the defect image characteristics which should be as discriminant as possible, to overcome the classification problems such as the inter-class similarity or intra-class difference, and help increasing the classification efficiency. Moreover, in the published works, the defect
* Zoheir Mentouri [email protected]; [email protected] 1
Research Center in Industrial Technologies – CRTI, P.O. Box 64, 16014 Cheraga, Algiers, Algeria
2
Lab. of Inverse Problems, Université 8 Mai 1945, BP 401, Guelma, Algeria
3
Lab. of Electrical Engineering, Université 8 Mai 1945, BP 401, Guelma, Algeria
detection has mostly been treated as a texture analysis issue [1], where two important families of techniques, namely the filtering and statistical approaches, have been very popular. The experienced techniques of spatial and joint spatial/ frequency domain
Data Loading...
