METAL'S MICROSTRUCTURE IMAGES CLASSIFICATION BY USING ARTIFICIAL NEURAL NETWORKS

Abstract

The objective of this study was to develop a back-propagation artificial neural network (ANN) model that could be distinguished the microstructure of metals instead of the traditional methods which are cost and need more time in addition some time give wrong results . Although the colour indices associated with image pixels were used as inputs, it was assumed that the ANN model could develop the ability to use other information, such as shapes that are implicit in these data. The 504x504 pixel images were taken in the field in standard situations and were then cropped to 10x10-pixel images depicting only one microstructure metal. The metals images with size 10x10 which is sufficient for recognizing by ANN. There are four images of different materials.