Detection of straw default by artificial vision

Auteurs :  S.Taleb, S.Ziani, S.Boulkroune
Année : 2018
Domaine : Electronique
Type : Communication
Conférence: Symposium on Materials Chemistry (ISyMC)
Résumé en PDF :  (résumé en pdf)
Fulltext en PDF :  (.pdf)
Mots clés : classification, histogram, attribute, probability, sheet.

Résumé : 

The aim of our work is the development of a detection method of straw default on the surface of a rolled steel sheet, in order to integrate it into a vision system. Our study is based on a bidimensional classification by two classes. It is divided into three parts: The classification, the detection of the presence of a defect and the recognition of this type of defect. We simulate the straw default with Matlab. First, we proceeded to a preliminary study to obtain the density functions and the priori probabilities of the sheet and the straw defect, and then we integrated them in the Bayes formula. Such as the histograms of sheet flawless and straw defect have an allure close to the Gaussian distribution, we made a statistical inference from a significant number of samples of these histograms and we estimate the parameters (expectation µ and standard deviation σ) of Normal distribution representatives of these density functions with method of the maximum-likelihood. Seen thatthese histograms overlap on a common interval, detecting the presence of a defect based only on the attribute of the intensity of grayscalewill be always with a probability of error. It was then necessary to add a second attribute, that of the spatial variabilityto minimize this error and ensure the reliability of the detection of presence of the defect. . Finally, to optimize the operation of straw defect detection, we introduced geometrical attributes of this defect (elongation criteria, dimensional criteria, criteria specific to the shape and area ratios). This method will subsequently be implemented in a program using image processing software adapted to a matrix camera. In conclusion, this work willcontribute to early detection of strawdefect which willincrease the quantity and the quality of the product, prevent the evolution of defects (and therefore the stops of the chainand rejection of the product) and simplify working conditions of the operator.