Featured image of Researchers Detect Defects in Metal 3D Printed Parts Using Neural Networks
This article is free for you and free from outside influence. To keep things this way, we finance it through advertising and shopping links. If you purchase using a shopping link, we may earn a commission. Learn more
Algorithm for the Win

Researchers Detect Defects in Metal 3D Printed Parts Using Neural Networks

Picture of Hanna Watkin
by Hanna Watkin
Published Sep 21, 2018
Advertisement

Researchers from Lawrence Livermore National Laboratory have developed a means to detect in real time whether a metal 3D printed part will be of good quality. Using machine learning algorithms (called convolutional neural networks) and a live video feed the team can spot defects in a piece as it's printing.