For many decades, ultrasonic imaging inspection has been adopted as a principal method to detect multiple defects, e.g., void and corrosion. However, the data interpretation relies on an inspector’s subjective judgment, thus making the results vulnerable to human error. Nowadays, advanced computer vision techniques reveal new perspectives on the high-level visual understanding of universal tasks. This research aims to develop an efficient automatic ultrasonic image analysis system for nondestructive testing (NDT) using the latest visual information processing technique. To this end, we first established an ultrasonic inspection image dataset containing 6849 ultrasonic scan images with full defect/no-defect annotations. Using the dataset, we performed a comprehensive experimental comparison of various computer vision techniques, including both conventional methods using hand-crafted visual features and the most recent convolutional neural networks (CNN) which generate multiple-layer stacking for representation learning. In the computer vision community, the two groups are referred to as shallow and deep learning, respectively. Experimental results make it clear that the deep learning-enabled system outperformed conventional (shallow) learning schemes by a large margin. We believe this benchmarking could be used as a reference for similar research dealing with automatic defect detection in ultrasonic imaging inspection.

The JapanSociety of Mechanical Engineers ing of four finger 7075 allo)fs. Different d]e materials: preheated toel steel (N 250eC), graphite and no+asbestos Sindanyo*. The solid-cation behavieur was etcamined by numerTcal finite element model. The moclel was to previde the temperature distribution during soEidificatien, to lecate the shrinkage petosity, and te estimate the proper ejection time of the semi-solid procluct. The results show thet thixoformed 7075 proclucts predueed by using Sindanyo die have the hlghest tensi]e properties due to the minim"m ef heat foss dur;ng solidification, resulting in cervv pletely die filling before solidification and the ending ef freexing at the gate. * Sindanyo grade H61 has a substr2te of hlgh Alumina cement w;th selected fihres to give goed strength at e]evated temperature.