期刊文献+
共找到2篇文章
< 1 >
每页显示 20 50 100
Microstructure and Optical Characterization of Magnetron Sputtered NbN Thin Films
1
作者 du xin-kang WANG Tian-min +2 位作者 WANG Cong CHEN Bu-liang ZHOU Long 《Chinese Journal of Aeronautics》 SCIE EI CAS CSCD 2007年第2期140-144,共5页
Some fundamental studies on the preparation, structure and optical properties of NbN films were carried out. NbN thin films were deposited by DC reactive magnetron sputtering at different N2 partial pressures and diff... Some fundamental studies on the preparation, structure and optical properties of NbN films were carried out. NbN thin films were deposited by DC reactive magnetron sputtering at different N2 partial pressures and different substrate temperatures ranging from -50℃ to 600℃. X-ray diffraction analysis (XRD) and scanning electron microscopy (SEM) were employed to characterize their phase components, microstructures, grain sizes and surface morphology. Optical properties inclusive of refractive indexes, extinction coefficients and transmittance of the NbN films under different sputtering conditions were measured. With the increase in the N2 partial pressure, 6-NbN phase structure gets forming and the grain size and lattice constant of the cubic NbN increasing. The deposited NbN film has relatively high values of refractive index and extinction coefficient in the wavelength ranging from 240 nm to 830 nm. Substrate temperature exerts notable influences on the microstructure and optical transmittance of the NbN films. The grain sizes of the 6-NbN film remarkably increase with the rise of the substrate temperature, while the transmittance of the films with the same thickness decreases. Ultra-fine granular film with particle size of several nanometers forms when the substrate is cooled to -50℃, and a remarkable augmentation of transmittance could be noticed under so low a temperature. 展开更多
关键词 reactive magnetron sputtering NbN thin film MICROSTRUCTURE optical properties
下载PDF
基于MobileNetV2特征的肝郁线智能诊断研究 被引量:2
2
作者 杜鑫康 张东晓 王彦晖 《中华中医药杂志》 CAS CSCD 北大核心 2021年第7期4291-4295,共5页
目的:解决智能舌诊中肝郁线自动识别问题。方法:设计了一个掩模,在使用深度卷积神经网络模型自动提取特征时,专注于肝郁线所在部位。在此特征基础上,搭建了一个前馈神经网络,用于识别肝郁线。通过随机调整舌体图像的方式增强数据,解决... 目的:解决智能舌诊中肝郁线自动识别问题。方法:设计了一个掩模,在使用深度卷积神经网络模型自动提取特征时,专注于肝郁线所在部位。在此特征基础上,搭建了一个前馈神经网络,用于识别肝郁线。通过随机调整舌体图像的方式增强数据,解决样本数量不平衡的问题。结果:掩模和数据增强可以大幅提升识别精确率,本研究所提模型的各项指标均优于经典分类算法。结论:在舌体图像清晰可辨的情况下,使用机器自动识别肝郁线是切实可行的。 展开更多
关键词 智能舌诊 肝郁线 数据增强 掩模
原文传递
上一页 1 下一页 到第
使用帮助 返回顶部