计算机科学 ›› 2018, Vol. 45 ›› Issue (11A): 180-184.

• 模式识别与图像处理 • 上一篇    下一篇

一种基于AKAZE算法的多视图几何三维重建方法

周泩朴, 耿国华, 李康, 王飘   

  1. 西北大学信息科学与技术学院 西安710127;
    西北大学文化遗产数字化国家地方联合工程研究中心 西安710127
  • 出版日期:2019-02-26 发布日期:2019-02-26
  • 通讯作者: 耿国华 女,博士,教授,主要研究方向为智能信息处理和文化遗产数字化保护等,E-mail:603395985@qq.com
  • 作者简介:周泩朴 男,硕士,主要研究方向为文化遗产数字化保护、大场景三维重建等,E-mail:zhoushengpu@126.com
  • 基金资助:
    本文受国家自然科学基金项目(61673319),陕西省重点研发计划(2017ZDCXL-GY-03-01-01),陕西省教育厅专项科研计划基金项目(16JK1775)资助。

Multi-view Geometric 3D Reconstruction Method Based on AKAZE Algorithm

ZHOU Sheng-pu, GENG Guo-hua, LI Kang, WANG Piao   

  1. School of Information Science and Technology,Northwest University,Xi’an 710127,China
    National-Local Joint Engineering Research Center of Cultural Heritage Digitization,Northwest University,Xi’an 710127,China
  • Online:2019-02-26 Published:2019-02-26

摘要: 针对增量式运动恢复结构算法在多视图几何三维重建算法中运行效率低的问题,提出了一种基于AKAZE算法的多视图几何三维重建方法。首先对利用摄像机获得的目标图像使用AKAZE算法检测特征并匹配,并使用随机抽取一致性算法和三视图约束剔除弱匹配图像。然后根据匹配图间的相对位姿参数,通过最小二乘法解算全局旋转参数,并利用三视图约束关系求解全局位移参数。最后进行一次光束法平差优化。实验结果表明,该算法在改善重建效果的基础上提高了处理效率,能够满足快速处理的需求。

关键词: AKAZE算法, 光束平差, 全局三维重建, 随机抽取一致性算法

Abstract: Aiming at the low efficiency of incremental motion recovery structure algorithm in multi-view geometric 3D reconstruction algorithm,a multi-view geometric 3D reconstruction method based on AKAZE algorithm was proposed.The target image obtained by the camera is detected and matched by AKAZE algorithm,and the weak matching image is eliminated by using the random sample consensus algorithm and the three view constraints.Then the global rotation parameters are solved by the least square method according to the relative position and attitude parameters of the matching graphs,and the global displacement parameters are solved by using the three-view constraint relation.Finally,the bundle adjustment optimization is carried out.The experimental results show that the proposed algorithm can improve the processing efficiency and meet the needs of fast processing on the basis of improving the reconstruction effect.

Key words: AKAZE algorithm, Bundle adjustment, Global 3D reconstruction, RANSAC algorithm

中图分类号: 

  • TP391.7
[1]CHENG X J,ZHANG H F,XIE R.Study on 3D laser scanning modeling method for Large-Scale history building[C]∥International Conference on Computer Application and System Mo-deling.IEEE,2010:V7-573-V7-577.
[2]GRACIÁ L,SAEZ-BARONA S,CARRIOÓN D,et al.A System for Real-Time Multi-View 3D Reconstruction[C]∥Workshops on Database and Expert Systems Applications.IEEE Computer Society,2010:235-239.
[3]SCHÖNING J,HEIDEMANN G.Evaluation of multi-view 3D reconstruction software[C]∥Computer Analysis of Images and Patterns.2015:450-461.
[4]HÄMING K,PETERS G.The structure-from-motion recon-struction pipeline-A survey with focus on short image sequences[J].Kybernetika-Praha,2010,5(5):926-937.
[5]SABZEVARI R,BUE A D,MURINO V.Structure from Motion and Photometric Stereo for Dense 3D Shape Recovery[M]// Image Analysis and Processing-ICIAP 2011.Springer Berlin Heidelberg,2011:660-669.
[6]SCHÖNBERGER J L,FRAHM J M.Structure-from-Motion Revisited[C]∥Computer Vision and Pattern Recognition.IEEE,2016.
[7]MOULON P,MONASSE P,MARLET R.Adaptive Structure from Motion with a Contrario,Model Estimation[M]∥Compu-ter Vision-ACCV 2012.Springer Berlin Heidelberg,2012:257-270.
[8]MOULON P,MONASSE P,MARLET R.Global Fusion of Re-lative Motions for Robust,Accurate and Scalable Structure from Motion[C]∥IEEE International Conference on Computer Vision.IEEE Computer Society,2013:3248-3255.
[9]ALCANTARILLA P F,BARTOLI A,DAVISON A J.KAZE Features[M]∥Computer Vision-ECCV 2012.Springer Berlin Heidelberg,2012:214-227.
[10]JIANG G,LIU L,ZHU W,et al.A 127 fps in full hd accelerator based on optimized AKAZE with efficiency and effectiveness for image feature extraction[C]∥IEEE Design Automation Con-ference.2015:1-6.
[11]CHEN Y,CHAN A B,LIN Z,et al.Efficient tree-structured SfM by RANSAC generalized Procrustes analysis[J].Computer Vision & Image Understanding,2017,157(C):179-189.
[12]许可乐.图像局部不变特征检测与描述技术研究[D].长沙:国防科学技术大学,2013.
[13]吴鹏,于秋则,闵顺新.一种快速鲁棒的SAR图像匹配算法[J].计算机科学,2017,44(7):283-288.
[14]SHI L M,GUO F S,HU Z Y,et al.An Improved PMVS through Scene Geometric Information[J].Acta Automatica Si-nica,2011,37(5):560-568.
[15]田歌,赵阳,张浩,等.基于Delaunay算法三角形网格划分的角点优化处理[C]∥北京力学会学术年会.2010.
[1] 赵人行, 徐频捷, 刘瑶.
基于深度卷积残差网络的心电单导联房颤检测方法
ECG-based Atrial Fibrillation Detection Based on Deep Convolutional Residual Neural Network
计算机科学, 2022, 49(5): 186-193. https://doi.org/10.11896/jsjkx.220200002
[2] 黄梅根, 刘川, 杜欢, 刘佳乐.
基于知识图谱的认知诊断模型及其在教辅中的应用研究
Research on Cognitive Diagnosis Model Based on Knowledge Graph and Its Application in Teaching Assistant
计算机科学, 2021, 48(6A): 644-648. https://doi.org/10.11896/jsjkx.200700163
[3] 王桂平, 刘君, 罗宪, 陈旺桥.
一个基于多种评判模式的在线评判系统
Online Judge System Based on Multiple Judgement Modes
计算机科学, 2020, 47(11A): 657-661. https://doi.org/10.11896/jsjkx.200500048
[4] 高强, 高敬阳, 赵地.
GNNI U-net:基于组归一化与最近邻插值的MRI左心室轮廓精准分割网络
GNNI U-net:Precise Segmentation Neural Network of Left Ventricular Contours for MRI Images Based on Group Normalization and Nearest Interpolation
计算机科学, 2020, 47(8): 213-220. https://doi.org/10.11896/jsjkx.190600026
[5] 岳笑含, 许晓健, 王溪波.
面向FMS基于改进的混合PSO-GA的多AGV调度算法研究
Research on Muti-AGV Sechduling Algorithm Based on Improved Hybrid PSO-GA for FMS
计算机科学, 2018, 45(11A): 167-171.
[6] 张泽中, 高敬阳, 吕纲, 赵地.
基于深度学习的胃癌病理图像分类方法
Pathological Image Classification of Gastric Cancer Based on Depth Learning
计算机科学, 2018, 45(11A): 263-268.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!