Surface defect classification of steel strip based on machine vision

Steel is a widely used material in the industry and household. The presence of defects on the surface of steel strip has serious implications and limit its use for quality purpose significantly. The primary objective of this paper is to develop a method for classification of steel surface defects such as Blister, Scratch and pseudo defect like Water droplet. This paper proposes artificial neural network based methodology to classify the defects. The neural network is responsible for making intelligent classification based on the training for various types of defects. Proposed approach has been experimentally found to be very effective.