Classification of NDT images with the Machine Consciousness Toolbox (MCT) - Part 2: Analysis of NDT images from multiple sources

This article deals with neural network classification of NDT images using t he Machine Consciousness Toolbox software. The analysis of NDT images from multiple sources is carried out and the output of the neural network is dis played in a graphical format as a reconstructed image and a decision-making image. A comparison of the effectiveness of feeding the neural network wit h single images (part 1) and multiple images in a pair-wise manner is made. The effect of increasing the number of connections per neurone on image an alysis is also addressed.