摘要
本文通过研究高分辨电子显微像的图像处理技术，系统地建立了两种高分辨原子像亮点中心精确定位的处理方法：灰度梯度检测法和峰谷提取－灰度平均法。基于原子像亮点中心定位技术，使用最小二乘法，建立了测量高分辨像中局部点阵参数和晶格畸变的实际处理过程。结合像模拟、图像匹配等手段，详细研究了从高分辨像中提取元素分布、原子结构等信息的定量分析方法；并建立了一套ＵＮＩＸ平台上的高分辨像定量分析程序包，具有较高的精度和广泛的用途。
In this paper, two procedures of the center localization for the bright spot in digitized HREM image have been developed; gray gradient detection and valleymeshextractingand grayaverage methods, Based on the techniques of the spotcenterlocalization, the procedures of measuring local lattice parameters and lattice displacement in HREM images have been built by use of the least square methods. Combining image simulation with image matching, the methods of quantitative analysis of HREM images have been studied for information such as element composition and atomic structures in detail. Written by C/C++ language, a program package with high precision and wide uses for analysis of HREM images has been developed under the UNIX environment.
出处
《电子显微学报》
CAS
CSCD
1998年第1期72-77,共6页
Journal of Chinese Electron Microscopy Society
基金
国家自然科学基金
国家高技术新材料领域专家委员会资助
关键词
电子显微镜
点阵畸变
定量分析
图像处理
HREM image\ \ lattice distortion\ \ quantitative analysis\ \ image processing
