Abstract
Due to the required high demands of quality products and to higher productivity rates, automated inspection systems have to be implemented in the industry. These inspection systems have to be reliable, fast, robust and flexible in their functionality in order to meet the high demands of the industry. An X-ray image of the product is taken and analysed by the system. The most important stages of the inspection process are the segmentation of the image into meaningful objects and the consequent analysis of these. This paper presents a Competitive Hopfield Neural Network (CHNN) for the segmentation of dual-band X-ray images of inspected product and a high-level detection of foreign bodies/defects based on fuzzy logic. Keywords: composite materials, image segmentation, Hopfield neural network, dual-band image, X-ray image, fuzzy logic