Machine learning to analyze images of shocked materials for precise andaccurate measurements

• Published in: Journal of applied physics 2017-09, Vol.122 (10)
• Main Authors: Marylesa, Howard, Hock, Margaret C, Meehan, B T, Ramos, Kyle J, Bolme, Cindy A, Sandberg, Richard L
• Format: Article
• Language: English
• Subjects: Adaptive algorithms; Applied physics; Artificial intelligence; Computer simulation; Discriminant analysis; Image segmentation; Machine learning; Pixels; Shock waves
• ISSN: 0021-8979; 1089-7550

A supervised machine learning algorithm, called locally adaptive discriminant analysis(LADA), has been developed to locate boundaries between identifiable image features thathave varying intensities. LADA is an adaptation of image segmentation, which includestechniques that find the positions of image features (classes) using statistical intensitydistributions for each class in the image. In order to place a pixel in the proper class,LADA considers the intensity at that pixel and the distribution of intensities in local(nearby) pixels. This paper presents the use of LADA to provide, with statisticaluncertainties, the positions and shapes of features within ultrafast images of shockwaves. We demonstrate the ability to locate image features including crystals, densitychanges associated with shock waves, and material jetting caused by shock waves. Thisalgorithm can analyze images that exhibit a wide range of physical phenomena because itdoes not rely on comparison to a model. LADA enables analysis of images from shock physicswith statistical rigor independent of underlying models or simulations.