Abstract：Based on the structure from motion (SFM), a 3D reconstruction and measurement method based on multi view stereo vision is proposed. Firstly, according to the unstable characteristics of the SFM method, the SFM solution process is analyzed and optimized by the bundle adjustment method. Secondly, the SFM reconstruction result is sparse point cloud, and the patch-based dense reconstruction algorithm is used to regenerate the dense 3D points. The cloud is then reconstructed by a Poisson algorithm for a 3D surface of a dense point cloud. Finally, the 3D size information of the model is obtained. The 3D measurement experiment of the sand pile at a construction site is carried out. The experimental results verify the effectiveness and feasibility of the method, improve the reconstruction ability and accuracy, and consider the actual measurement error and reconstruction error of the target, which can meet the requirements of actual intelligent measurement applications.
陈辉,杨剑,黄晓铭,裴琴,黄梅丽. 基于多视图立体视觉的沙堆三维尺寸测量研究[J]. 计量学报, 2019, 40(3): 403-408.
CHEN Hui,YANG Jian,HUANG Xiao-ming,PEI Qin,HUANG Mei-li. Research on 3D Measurement of Sand Pile Based on Multi View Stereo Vision. Acta Metrologica Sinica, 2019, 40(3): 403-408.
［1］程淑红, 刘洁. 基于多尺度韦伯脸和梯度脸的复杂光照下人脸识别研究 ［J］. 计量学报, 2017, 38(1): 60-64.
Cheng S H, Liu J. Face recognition under complex illumination based on multi-scale Weber-face and Gradient-face ［J］. Acta Metrologica Sinica, 2017, 38(1): 60-64.
［2］韦争亮, 古耀达, 黄志斌, 等. 双目立体视觉中特征点三维坐标重构校准研究 ［J］. 计量学报, 2014, 35(2):102-107.
Wei Z L, Gu Y D, Huang Z B, et al. Research on calibration of three dimensional coordinate reconstruction of feature points in binocular stereo vision ［J］. Acta Metrologica Sinica, 2014, 35(2): 102-107.
［3］Pollefeys M, Van Gool L. Visual modelling: From images to images ［J］. The Journal of Visualization and Computer Animation, 2002, 13(4): 199-209.
［4］Furukawa Y, Ponce J. Carved visual hulls for image-based modeling ［C］ // European Conference on Computer Vision. Berlin, Germany, 2006:564-577.
［5］Hornung A, Kobbelt L. Hierarchical volumetric multi-view stereo reconstruction of manifold surfaces based on dual graph embedding ［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA, 2006:503-510.
［6］Strecha C, Fransens R, Van Gool L. Combined depth and outlier estimation in multi-view stereo ［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, USA, 2006:2394-2401.
［7］Bradley D, Boubekeur T, Heidrich W. Accurate multi-view reconstruction using robust binocular stereo and surface meshing ［C］// IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Anchorage, USA, 2008:1-8.
［8］Paris S, Sillion F X, Quan L. A surface reconstruction method using global graph cut optimization ［J］. International Journal of Computer Vision, 2006, 66(2): 141-161.
［9］ Esteban C H, Schmitt F. Silhouette and stereo fusion for 3d object modeling ［J］. Computer Vision and Image Understanding, 2004, 96(3): 367-392.
［10］陈辉, 马世伟, Andreas N. 基于激光扫描和SFM的非同步点云三维重构方法 ［J］. 仪器仪表学报, 2016, 17(5): 1148-1157.
Chen H, Ma S W, Andreas N. Non-synchronous point cloud algorithm for 3D reconstruction based on laser scanning and SFM ［J］. Chinese Journal of Scientific Instrument, 2016, 17(5):1148-1157.
［11］Furukawa Y,Ponce J. Accurate, dense, and robust multiview stereopsis ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(8): 1362-1376.
［12］史利民, 郭复胜, 胡占义. 利用空间几何信息的改进PMVS算法 ［J］. 自动化学报, 2011, 37(5): 560-568.
Shi L M, Guo F S, Hu Z Y. An improved PMVS through scene geometric information ［J］. Acta Automatica Sinica, 2011, 37(5): 560-568.
2019, Vol. 40 
