The influence of lighting on the detection of straw defects by artificial vision

Auteurs :  S.Taleb, S.Ziani, S.Boulkroune, K.Slimani
Année : 2018
Domaine : Electronique
Type : Communication
Conférence: Optics and Photonics Algeria 2018(OPAL 2018)
Résumé en PDF :  (résumé en pdf)
Fulltext en PDF : 
Mots clés : Lighting, classification, machine vision, straw.

Résumé : 

The aim of this work is to study the influence of lighting on a machine vision system destined for the detection of surface defects on rolled sheets in a cold rolling mill. A machine vision system consists of a camera, an acquisition and image processing card and a lighting device. This last is an essential element for the success of an artificial vision control system. In this project, we are particularly interested by the influence of lighting on the detection of straw defects on a rolled sheet. Our method is based on a statistical approach essentially using the Bayesian concept. It is divided into three parts: The classification, the detection of the presence of a defect and the recognition of the type of this defect. We have applied this method to straw defects that have undergone two types of lighting simulated with Matlab: Dark field lighting and coaxial lighting. A comparative study of these results allowed us to choose the dark field lighting because it gave conclusive results.