首页 > 论文 > 应用激光 > 37卷 > 6期(pp:881-887)

基于高斯过程回归与马尔科夫随机场的三维深度图像重构算法

3D Reconstruction Algorithm Based on Gaussian Process Regression and Markov Random Field

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

由于激光雷达点云数据有无序性、稀疏性和信息量有限的问题, 提出了一种能够将点云数据与对应图像进行三维图像重建的算法, 该方法首先将点云数据体素化, 利用点特征直方图有效地选择深度点进行标记并消除体素中的异常点; 针对传统插值方法估计精度低的缺陷, 利用高斯过程回归方法强大的非线性拟合能力和小样本学习能力, 提高了内插点估计精度, 获得稠密点云; 最后利用马尔科夫随机场对图像灰度数据和三维插值点进行融合来构建三维深度图。定性定量仿真实验结果表明, 提出的算法大大提升了三维重建的鲁棒性与重构精度, 可用于复杂路况中的无人驾驶应用。

Abstract

Since the point-cloud data has randomness, sparsity and the limited information in laser radar, a novel and robust 3D image reconstruction algorithm based on depth point-cloud and its corresponding image in this paper. The proposed algorithm firstly makes the point-cloud data voxelization, which adopts the point feature histogram(PFH)to effectively choose the depth-point and remove abnormal point in voxel; aiming at the shortcomings of the traditional interpolation method, the Gaussian process regression method(GPF) is used to improve the accuracy of the interpolation point and obtain the dense point-cloud; finally, the Markov random field is adopted to merge the gray data and points-cloud so as to build 3D depth image. The results of qualitative and quantitative simulation show that proposed algorithm is superior to other existing algorithms in terms of RMSE and run time.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TN958.98;TP391

DOI:10.14128/j.cnki.al.20173705.881

基金项目:上海市教育发展基金会“晨光计划”资助项目(项目编号: 12CGB22); 高等学校访问学者专业发展资助项目(项目编号: FX2012122)

收稿日期:2017-03-07

修改稿日期:2017-11-02

网络出版日期:--

作者单位    点击查看

吴倩倩:长春理工大学 光电信息学院, 吉林 长春 130000
蔡艳:澳门科技大学 资讯科技学院, 澳门999078南昌航空大学 测试与光电工程学院, 江西 南昌330000

联系人作者:吴倩倩(27609267@qq.com)

备注:吴倩倩 (1984-), 女, 讲师, 主要从事光学工程等相关研究。

【1】刘昱岗, 王卓君, 王福景, 等.基于双目立体视觉的倒车环境障碍物测量方法[J].交通运输系统工程与信息, 2016, 16(4): 79-87.

【2】杨景豪, 刘巍, 刘阳, 等.双目立体视觉测量系统的标定[J].光学 精密工程, 2016, 24(2): 300-308.

【3】曾日金.基于双目立体视觉的障碍物检测系统研究与实现[D].桂林: 桂林电子科技大学, 2014.

【4】韩硕, 童晓冲, 卢学良, 等.手持线激光扫描视频三维重建中的运动线提取算法[J].计算机应用与软件, 2016, 33(7): 154-158..

【5】DIEBEL J, THRUN S.An application of Markov random elds to range sensing[J].Advances in Neural Information Processing Systems, 2005(18): 291-298.

【6】ZHU J, WANG L, GAO J, et al.Spatial-temporal fusion for high accuracy depth maps using dynamic MRFs[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(5): 899-909.

【7】LU J, MIN D, PAHWA R S, et al.A revisit to mrf-based depth map super-resolution and enhancement: Acoustics, IEEE International Conference on Speech and Signal Processing (ICASSP), 2011[C].[S.I.]:[s.n.], 2011: 985-988.

【8】MURPHY K P, WEISS Y, JORDAN M I.Loopy belief propagation for approximate inference: An empirical study. Proceedings of the Fifteenth Conference on Uncertainty in Articial Intelligence[C]. Morgan Kaufmann Publishers Inc., 1999: 467-475.

【9】PARK J, KIM H, TAI Y W, et al. High quality depth map upsampling: IEEE International Conference on Computer Vision (ICCV)[C]. IEEE, 2011: 1623-1630.

【10】YANG Q, YANG R, DAVIS J, et al. Spatial-depth super resolution for range images: Computer Vision and Pattern Recognition CVPR 2007, IEEE Conference on[C]. IEEE, 2007: 1-8.

【11】CHAN D, BUISMAN H, THEOBALT C, et al.A noise-aware lter for real-time depth upsampling: Workshop on Multi-camera and Multi-modal Sensor Fusion Algorithms and Applications-M2SFA2 2008[C].

【12】DOLSON J, BAEK J, PLAGEMANN C, et al.Upsampling range data in dynamic environments: Computer Vision and Pattern Recognition(CVPR), 2010 IEEE Conference on[C].IEEE, 2010: 1141-1148.

【13】FOIX S, ALENYA G, TORRAS C.Lock-in time-of-ight (tof) cameras: a survey[J].Sensors Journal IEEE, 2011, 11(9): 1917-1926.

【14】HARRISON A, NEWMAN P.Image and sparse laser fusion for dense scene reconstruction[M].Field and Service Robotics Springer, 2010: 219-228.

【15】SCHOENBERG J R, NATHAN A, CAMPBELL M.Segmentation of dense range information in complex urban scenes: Intelligent Robots and Systems(IROS), 2010 IEEE/RSJ International Conference on[C].IEEE, 2010: 2033-2038.

【16】张爱武, 李文宁, 段乙好, 等.结合点特征直方图的点云分类方法[J].计算机辅助设计与图形学学报, 2016, 28(5): 795-801.

【17】FERSTL D, REINBACHER C, BISCHOF H.Image guided depth up sampling using anisotropic total generalized variation: 2013 IEEE International Conference on Computer Vision(ICCV)[C].[S.I.]:[s.n.], 2013: 993-1000.

【18】GRINGARTEN E, DEUTSCH C V.Teacher’s aide variogram interpretation and modeling[J].Mathematical Geology, 2001, 33(4): 507-534.

引用该论文

Wu Qianqian,Cai Yan. 3D Reconstruction Algorithm Based on Gaussian Process Regression and Markov Random Field[J]. APPLIED LASER, 2017, 37(6): 881-887

吴倩倩,蔡艳. 基于高斯过程回归与马尔科夫随机场的三维深度图像重构算法[J]. 应用激光, 2017, 37(6): 881-887

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF