Spherical-angular dark field imaging and sensitive microstructural phase clustering with unsupervised machine learning

Processing Laue Microdiffraction Raster Scanning Patterns with Machine Learning Algorithms: A Case Study with a Fatigued Polycrystalline Sample

“…In analogy to the preceding work [31,32,35,38], the treatment consisted of two steps: (i) feature extraction and (ii) grid segmentation. For pragmatic consideration, we adopted unsupervised machine learning algorithms since the training data might not be always available.…”

“…Another candidate for factorization is non-negative matrix factorization (NMF) which has been used in processing electron microscopy images [35,38]. In our practice, we found that NMF costed much more computation time than PCA.…”

“…To begin with, each image needed to be normalized to eliminate systematic errors. Normalization was accomplished by subtracting the mean gray level from the gray level and dividing by the standard deviation of the gray levels, in accordance with McAuliffe et al [35].…”

SIGMA: Spectral Interpretation Using Gaussian Mixtures and Autoencoder

“…With autoencoders, the dimensionality of the data can be reduced without losing essential features. Despite being widely employed in other types of HSI data (Lin et al., 2013; C. Tao et al., 2015; X. Tao et al., 2022) and some electron microscopic data (Ede, 2020; McAuliffe et al., 2020), autoencoders have had limited applications in HSI‐EDS data.…”

Desenvolvimento de uma ferramenta mobile para teste Elisa em microcircuitos e detecção de cores com inteligência artificial

“…O principal objetivo do clustering é categorizar dados semelhantes com base em medidas de similaridade 79 . Clustering ou agrupamento é definido como aprendizado de máquina não supervisionado e permite explorar a estrutura dos conjuntos de dados identificando informações antes difíceis de se interpretar 80 .…”