基于数码体视显微镜的微小物体三维重建
PDF下载 (426)刘 奕 1,郁 梅 1*,崔 力 1,王一刚 1,2,范胜利 1,2.基于数码体视显微镜的微小物体三维重建[J].宁波大学学报(理工版),2015,28(02):42-47.DOI:
LIU Yi 1,YU Mei 1*,CUI Li 1,WANG Yi-gang 1,2,Fan Sheng-li 1,2.New Micro Object Reconstruction Method Based on Digital Stereo Microscope[J].Journal of Ningbo University(Natural Science & Engineering Edition),2015,28(02):42-47.DOI:
| Title: | New Micro Object Reconstruction Method Based on Digital Stereo Microscope |
| 作者: | 刘 奕 1, 郁 梅 1*, 崔 力 1, 王一刚 1, 2, 范胜利 1, 2 |
| Author(s): | LIU Yi 1, YU Mei 1*, CUI Li 1, WANG Yi-gang 1, 2, Fan Sheng-li 1, 2 |
| 关键词: | 体视显微镜; 平行双目系统; 三维重建; 参数标定 |
| Keywords: | stereo microscope; parallel binocular system; three-dimensional reconstruction; parameter calibration |
| 分类号: | TN911.7 |
| 文献标识码: | A |
| 摘要: | 将显微双目系统看作一个整体, 提出了基于极线校正的体视显微镜三维重建方法. 根据平行双目成像原理建立简易的三维重建模型, 实现了对实验物体的三维坐标恢复. 首先, 采用近似欧式极线校正算法校正原始拍摄立体标定图像, 并只需标定2个三维重建参数; 然后用非局部代价聚合匹配算法匹配立体图像得到视差值; 最后根据立体成像几何法得到物体三维重建坐标. 实验结果表明: X和Y方向测量误差为3%, Z方向测量误差为3.2%~9.1%, 故在允许误差下可用于测量微小物体的三维尺寸. |
| Abstract: | Due to the difficult and cumbersome camera parameter calibration process in stereo microscopic image reconstruction and the vertical displacement in camera’s imaging of the same object caused by small deviation of actual stereo microscope’s parallel optic path, a new three-dimensional reconstruction method for stereo microscopic image is proposed in this paper, where epipolar rectification is used by regarding microscope binocular system as a whole. The proposed method uses three-dimensional reconstruction model to recover three-dimensional coordinates of test object with considering the characteristics of parallel binocular imaging principle. To obtain rectified stereo images and calibrate two three-dimension reconstruction parameters, Quasi-Euclidean epipolar rectification algorithm is utilized into original stereo image. Then, disparity map is obtained by using non-local cost aggregation matching algorithm on rectified stereo image. Finally, according to three-dimensional imaging geometry technique, the coordinates of three-dimension reconstruction are recovered. Experimental results show that for the proposed method, its measurement error in the directions X and Y is about 3% and while one in the direction Z is 3.2%-9.1%, which indicates that the proposed method can be used to measure three-dimensional size of micro object within tolerance error. |
| 参考文献 /References: | [1] Reingruber H, Zankel A, Mayrhofer C, et al. Quantitative characterization of microfiltration membranes by 3D reconstruction[J]. Journal of Membrane Science, 2011, 372(1):66-74. [2] Yagi T, Shinoura S, Umeda Y, et al. Surgical rationaliz- ation of living donor liver transplantation by abolition of hepatic artery reconstruction under a fixed microscope[J]. Clinical Transplantation, 2012, 26(6):877-883. [3] Cardone G, Yan X, Sinkovits R S, et al. Three- dimensional reconstruction of icosahedral particles from single micrographs in real time at the microscope[J]. Journal of Structural Biology, 2013, 183(3):329-341. [4] 王娜, 周权. 光学显微镜下微小物体的三维重构[J]. 光电工程, 2010, 37(11):84-90. [5] Gallo A, Muzzupappa M, Bruno F. 3D reconstruction of small sized objects from a sequence of multi-focused images[J]. Journal of Cultural Heritage, 2014, 15(2):173-182. [6] Schreier H W, Garcia D, Sutton M A. Advances in light microscope stereo vision[J]. Experimental Mechanics, 2004, 44(3):278-288. [7] Li K, Wang Q, Wu J, et al. Calibration error for dual-camera digital image correlation at microscale[J]. Optics and Lasers in Engineering, 2012, 50(7):971-975. [8] Wang Y, Liu J. 3D shape reconstruction of micro- structures via micro stereovision[C]//International Con- ference on Mechatronics and Automation, 2009:1861- 1865. [9] Li Zhongwei, Li Y F. Microscopic photometric stereo: A dense microstructure 3D measurement method[C]//IEEE International Conference on Robotics and Automation (ICRA), 2011:6009-6014. [10] 高宏伟. 计算机双目立体视觉[M]. 北京: 电子工业出版社, 2012:131. [11] Fusiello A, Irsara L. Quasi-euclidean epipolar rectifi- cation of uncalibrated images[J]. Machine Vision and Applications, 2011, 22(4):663-670. [12] Bay H, Tuytelaars T, Van Gool L. Surf: Speeded up robust features[M]. Berlin: Springer, 2006:404-417. [13] Muja M, Lowe D G. Fast approximate nearest neighbors with automatic algorithm configuration[C]//International Conference on Computer Vision Theory and Applications, 2009:331-340. [14] Fischler M A, Bolles R C. Random sample consensus: A paradigm for model fitting with applications to image analysis and automated cartography[J]. Communications of the ACM, 1981, 24(6):381-395. [15] Yang Q. A non-local cost aggregation method for stereo matching[C]//Conference on Computer Vision and Pattern Recognition, 2012:1402-1409. |
| 备注/Memo: | 收稿日期: 2014－03－31. 宁波大学学报（理工版）网址: http://journallg.nbu.edu.cn/基金项目: 国家自然科学基金（61271270）; 浙江省自然科学基金（Y1101240）; 宁波市自然科学基金（2011A610197, 2011A610200）.第一作者: 刘奕（1990－）, 男, 湖南株洲人, 在读硕士研究生, 主要研究方向: 数字音视频技术. E-mail: lewis081@126.com*通信作者: 郁梅（1968－）, 女, 江苏无锡人, 博士/教授, 主要研究方向: 视频通信及图像处理. E-mail: yumei@nbu.edu.cn 宁波大学学报（理工版）网址：http://journallg.nbu.edu.cn/ |
