基于改进SFM的三维重建算法研究
|
摘要
针对现有运动恢复结构算法重建模型存在点云稀疏等问题,提出一种利用不同匹配数据进行模型重建的算法。首先通过对比上下文直方图(CCH)生成匹配数据,利用M估计抽样一致(MSAC)估算图像基础矩阵,进而分解得到平移和旋转矩阵,并根据相机内参计算投影矩阵,然后利用KLT匹配算法更新匹配数据,最后三角化生成三维点云。该算法匹配精度高,图像基础矩阵易于收敛,通过位移实现特征点匹配,弥补了图像低频区域匹配数据不足的缺陷。实验结果表明,与现有算法相比,该算法生成的点云更致密;在真实环境下,该算法可用于物体三维重建。
Aiming at the sparse problem of object point cloud based on structure from motion method, a 3 D reconstruction method using different matching data is proposed. The matching points are calculated by contrast context histogram( CCH) algorithm. The M-estimation sampling consensus( MSAC) algorithm is used to calculate the fundamental matrix, the translation and rotation matrix are decomposed from fundamental matrix. The image projection matrix is obtained combining the camera internal parameters. KLT algorithm is used to update the matching data, and the point cloud is generated by triangulation principle. This method makes use of the advantage of high accuracy of CCH algorithm to make the calculation results of the basic matrix converge. Using KLT algo-rithm to realize the matching by displacement instead of description vector, it makes up for the deficiency of matching data in low frequency region. The experimental results show that the proposed algorithm is effective and feasible, and the reconstructed point cloud has advantages in comparison with existing algorithms, it can be used for building 3 D model of objects in the real scene.
Aiming at the sparse problem of object point cloud based on structure from motion method, a 3 D reconstruction method using different matching data is proposed. The matching points are calculated by contrast context histogram( CCH) algorithm. The M-estimation sampling consensus( MSAC) algorithm is used to calculate the fundamental matrix, the translation and rotation matrix are decomposed from fundamental matrix. The image projection matrix is obtained combining the camera internal parameters. KLT algorithm is used to update the matching data, and the point cloud is generated by triangulation principle. This method makes use of the advantage of high accuracy of CCH algorithm to make the calculation results of the basic matrix converge. Using KLT algo-rithm to realize the matching by displacement instead of description vector, it makes up for the deficiency of matching data in low frequency region. The experimental results show that the proposed algorithm is effective and feasible, and the reconstructed point cloud has advantages in comparison with existing algorithms, it can be used for building 3 D model of objects in the real scene.
引文
[1]佟帅,徐晓刚,易成涛,等.基于视觉的三维重建技术综述[J].计算机应用研究,2011,28(7):2411-2417.
[2]CAO M,CAO L,JIA W,et al.Evaluation of local features for structure from motion[J].Multimedia Tools&Applications,2018,77(9):1-15.
[3]WU C.Towards linear-time incremental structure from motion[C].International Conference on 3D Vision,IEEEComputer Society,2013:127-134.
[4]ZACH C.Robust bundle adjustment revisited[C].European Conference on Computer Vision.Springer,2014:772-787.
[5]ZHANG G,LIU H,DONG Z,et al.Efficient non-consecutive feature tracking for robust structure from motion[J].IEEETrans Image Process,2015,25(12):5957-5970.
[6]SHI J,TOMASI C.Good features to track[C].Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2002:593-600.
[7]LOWE D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[8]BAY H,ESS A,TUYTELAARS T,et al.Speeded-up robust features[J].Computer Vision&Image Understanding,2008,110(3):404-417.
[9]刘欣,李校林,谢灿,等.一种基于仿射变换的SURF图像配准算法[J].电子技术应用,2014,40(6):130-132.
[10]胡影峰.基于SIFT特征匹配算子的三维重建方法研究[J].自动化仪表,2011,32(1):30-32.
[11]景子君.运动法三维重建的研究与实现[D].合肥:中国科学技术大学,2018.
[12]阚江明.基于计算机视觉的活立木三维重建方法[D].北京:北京林业大学,2008.
[13]HARTLEY R I.In defence of the 8-point algorithm[C].International Conference on Computer Vision,IEEE Computer Society,1995:1064-1070.
[14]HUANG C R,CHEN C S,CHUNG P C.Contrast context histogram a discriminating local descriptor for image matching[C].International Conference on Pattern Recognition.IEEE,2006:53-56.
[15]HUANG C R,CHEN C S,CHUNG P C.Contrast context histogram-An efficient discriminating local descriptor for object recognition and image matching[J].Pattern Recognition,2008,41(10):3071-3077.
[16]STRECHA C,HANSEN W V,GOOL L V,et al.On benchmarking camera calibration and multi-view stereo for high resolution imagery[C].2018 IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2008:1-8.
[2]CAO M,CAO L,JIA W,et al.Evaluation of local features for structure from motion[J].Multimedia Tools&Applications,2018,77(9):1-15.
[3]WU C.Towards linear-time incremental structure from motion[C].International Conference on 3D Vision,IEEEComputer Society,2013:127-134.
[4]ZACH C.Robust bundle adjustment revisited[C].European Conference on Computer Vision.Springer,2014:772-787.
[5]ZHANG G,LIU H,DONG Z,et al.Efficient non-consecutive feature tracking for robust structure from motion[J].IEEETrans Image Process,2015,25(12):5957-5970.
[6]SHI J,TOMASI C.Good features to track[C].Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2002:593-600.
[7]LOWE D G.Distinctive image features from scale-invariant keypoints[J].International Journal of Computer Vision,2004,60(2):91-110.
[8]BAY H,ESS A,TUYTELAARS T,et al.Speeded-up robust features[J].Computer Vision&Image Understanding,2008,110(3):404-417.
[9]刘欣,李校林,谢灿,等.一种基于仿射变换的SURF图像配准算法[J].电子技术应用,2014,40(6):130-132.
[10]胡影峰.基于SIFT特征匹配算子的三维重建方法研究[J].自动化仪表,2011,32(1):30-32.
[11]景子君.运动法三维重建的研究与实现[D].合肥:中国科学技术大学,2018.
[12]阚江明.基于计算机视觉的活立木三维重建方法[D].北京:北京林业大学,2008.
[13]HARTLEY R I.In defence of the 8-point algorithm[C].International Conference on Computer Vision,IEEE Computer Society,1995:1064-1070.
[14]HUANG C R,CHEN C S,CHUNG P C.Contrast context histogram a discriminating local descriptor for image matching[C].International Conference on Pattern Recognition.IEEE,2006:53-56.
[15]HUANG C R,CHEN C S,CHUNG P C.Contrast context histogram-An efficient discriminating local descriptor for object recognition and image matching[J].Pattern Recognition,2008,41(10):3071-3077.
[16]STRECHA C,HANSEN W V,GOOL L V,et al.On benchmarking camera calibration and multi-view stereo for high resolution imagery[C].2018 IEEE Conference on Computer Vision and Pattern Recognition.IEEE,2008:1-8.