基于双目视觉的FIB-SEM 双束系统图像三维重建研究
端 方，张海刚，孙 巍，朱建鹏，王贤浩，邱婷婷，王 英*，杨明来*
基于双目视觉的FIB-SEM 双束系统图像三维重建研究
端 方，张海刚，孙 巍，朱建鹏，王贤浩，邱婷婷，王 英*，杨明来*
（1.上海应用技术大学轨道交通学院，上海 201418；2.上海交大-平湖智能光电研究院, 浙江 嘉兴314299；3.吉林农业大学信息技术学院，吉林 长春 130118；4.上海交通大学先进电子材料与器件平台，上海 200240；5.卡尔蔡司（上海）管理有限公司，上海 200131）
摘 要 由FIB与SEM结合而成的双束系统拥有高精度微纳加工能力的同时也具有超高空间分辨率成像功能。在微纳尺度下使用双目立体视觉进行三维重建对指导微纳材料加工等具有重要意义。本文提出了一种基于双目视觉的FIB-SEM 双束系统图像三维重建方案：利用双目视觉结合图像处理技术测量出样品表面的深度信息。实验结果表明，该方法显著提高了微纳样品表面深度信息的精准度，为SEM图像三维构建提供了新的思路和方法，为FIB-SEM系统的精细加工提供了有力的保障。
关键词 FIB-SEM双束系统；三维重建；双目立体视觉；数字图像处理
中图分类号：TP317.4；TB302
文献标识码：A doi:10.3969/j.issn.1000-6281.2024.06.011
Research on 3D reconstruction of FIB-SEM dual beam system image based on binocular vision
DUAN Fang1，ZHANG Haigang1，SUN Wei1，ZHU Jianpeng1，WANG Xianhao5，QIU Tingting5，WANG Ying 2，4*，YANG Minglai1，3*
(1．School of Railway Transportation, Shanghai Institute of Technology, Shanghai 201418；2. SJTU-Pinghu Institute of Intelligent Optoelectronics, Jiaxing Zhejiang 314299；3. College of Information Technology Jilin Agricultural University, Changchun Jilin 130118；4.Center for Advanced Electronics Materials and Devices, Shanghai Jiao Tong University, Shanghai 200240；5. ZEISS Research Microscopy Solutions, Carl Zeiss (Shanghai) Co. Ltd., Shanghai 200131，China)
Abstract The dual beam system, combining FIB and SEM, offers high-precision micro-nano processing and ultra-high spatial resolution imaging. The application of binocular stereo vision for 3D reconstruction at the micro-nano scale is crucial for guiding the processing of micro-nano materials. This paper proposed a 3D reconstruction scheme for FIB-SEM double beam system image based on binocular vision. Depth information of the sample surface was measured using binocular vision combined with image processing technology. Experimental results demonstrated that this method significantly enhanced the accuracy of surface depth information for micro and nano samples, providing a novel approach to 3D construction of SEM images and a robust foundation for precise processing with the FIB-SEM system.
Keywords FIB-SEM dual beam system; 3D reconstruction; Binocular stereo vision; digital image processing