首页|Southwest Petroleum University Reports Findings in Machine Learning (Phase divis ion and recognition of crystal HRTEM images based on machine learning and deep l earning)
Southwest Petroleum University Reports Findings in Machine Learning (Phase divis ion and recognition of crystal HRTEM images based on machine learning and deep l earning)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New research on Machine Learning is th e subject of a report. According to news reporting originating from Chengdu, Peo ple's Republic of China, by NewsRx correspondents, research stated, "The High Re solution Transmission Electron Microscope (HRTEM) images provide valuable insigh ts into the atomic microstructure, dislocation patterns, defects, and phase char acteristics of materials. However, the current analysis and research of HRTEM im ages of crystal materials heavily rely on manual expertise, which is labor-inten sive and susceptible to subjective errors." Our news editors obtained a quote from the research from Southwest Petroleum Uni versity, "This study proposes a combined machine learning and deep learning appr oach to automatically partition the same phase regions in crystal HRTEM images. The entire image is traversed by a sliding window to compute the amplitude spect rum of the Fast Fourier Transform (FFT) in each window. The generated data is tr ansformed into a 4-dimensional (4D) format. Principal component analysis (PCA) o n this 4D data estimates the number of feature regions. Non-negative matrix fact orization (NMF) then decomposes the data into a coefficient matrix representing feature region distribution, and a feature matrix corresponding to the FFT magni tude spectra. Phase recognition based on deep learning enables identifying the p hase of each feature region, thereby achieving automatic segmentation and recogn ition of phase regions in HRTEM images of crystals. Experiments on zirconium and oxide nanoparticle HRTEM images demonstrate the proposed method achieve the con sistency of manual analysis."
ChengduPeople's Republic of ChinaAsi aCyborgsEmerging TechnologiesMachine Learning
NETL
NSTL
