Applying deep learning in automatic and rapid measurement of lattice spacings in HRTEM images
Sign Up to like & getrecommendations! Published in 2020 at "Science China Materials"
DOI: 10.1007/s40843-020-1368-7
Abstract: 在图像处理与分析领域, 深度学习发挥着日渐重要的作用. 我们将深度学习应用于高分辨率透射电子显微镜(HRTEM)图像晶 面间距的自动快速测量中, 通过对随机样本数据训练得到具有U型 结构的神经网络, 开发了一种新的图像处理方法. 本方法能够自动 提取快速傅里叶变换(FFT)图像中的衍射斑点, 进一步在计算机视 觉技术的协助下, 可以自动计算与FFT图像中被识别的衍射点相对 应的晶格间距, 并与标准晶体结构数据进行比较. 以Fe3 O4 纳米粒子 的HRTEM图像为例, 用本方法进行自动测量的晶格间距与手动测 量的晶格间距相比, 误差小于1%. 我们的工作证明了深度学习技术 在协助晶体材料发展方面的巨大潜力. read more here.
Keywords: learning automatic; hrtem; deep learning; rapid measurement ... See more keywords
Automatic and rapid whole-body 3D shape measurement based on multinode 3D sensing and speckle projection.
Sign Up to like & getrecommendations! Published in 2017 at "Applied optics"
DOI: 10.1364/ao.56.008759
Abstract: Automatic and rapid whole-body 3D shape measurement has attracted extensive attention in recent years and been widely used in many fields. Rapid 3D reconstruction, automatic 3D registration, and dedicated system layout are critical factors to… read more here.
Keywords: body shape; rapid whole; measurement; whole body ... See more keywords