计算机应用 ›› 2016, Vol. 36 ›› Issue (2): 541-545.DOI: 10.11772/j.issn.1001-9081.2016.02.0541

• 虚拟现实与数字媒体 • 上一篇    下一篇

基于不确定度评价的Kinect深度图预处理

余亚玲, 张华, 刘桂华, 史晋芳   

  1. 西南科技大学 信息工程学院, 四川 绵阳 621010
  • 收稿日期:2015-07-01 修回日期:2015-10-05 出版日期:2016-02-10 发布日期:2016-02-03
  • 通讯作者: 余亚玲(1990-),女,四川内江人,硕士研究生,主要研究方向:图像处理、机器视觉。
  • 作者简介:张华(1969-),男,四川绵阳人,教授,博士,主要研究方向:机器学习、机器人;刘桂华(1972-),女,四川绵阳人,教授,博士,主要研究方向:图像处理、机器视觉;史晋芳(1977-),女,四川绵阳人,副教授,硕士,主要研究方向:智能化测控、图像处理。
  • 基金资助:
    四川省科技创新苗子工程培育项目(2015024);西南科技大学研究生创新基金资助项目(14ycx103);2014四川省科技支撑计划项目(2014GZ0021);四川省教育厅重点项目(14ZA0090);特殊环境机器人技术四川省重点实验室开放基金资助项目(13zxtk05)。

Kinect depth map preprocessing based on uncertainty evaluation

YU Yaling, ZHANG Hua, LIU Guihua, SHI Jinfang   

  1. School of Information Engineering, Southwest University of Science and Technology, Mianyang Sichuan 621010, China
  • Received:2015-07-01 Revised:2015-10-05 Online:2016-02-10 Published:2016-02-03

摘要: 针对应用在机器人三维(3D)场景感知测量中,Kinect深度图的联合双边滤波(JBF)存在降低原始场景深度信息精确度的制约性问题,提出一种新的预处理算法。首先,通过构建深度图的测量和采样模型,得到深度图的蒙特卡罗不确定度评价模型;其次,依据该模型计算得到深度值估计区间,实现噪声点与非噪声点的判定及滤除;最后,利用估计区间均值完成噪声点的修复。实验结果表明,该算法在噪声滤波的同时保证了非噪声的不变性;非噪声的不变性以及基于估计均值的噪声修复使原始深度梯度具有不变性;与联合彩色深度图的双边滤波相比,预处理结果图物体边缘轮廓清晰不变且其均方误差降低了15.25%~28.79%。因此,该预处理算法达到了提高三维场景深度信息精确度的目的。

关键词: 深度图, 蒙特卡罗模型, 不确定度评价, 噪声滤波, 深度复原

Abstract: A new Kinect depth map pretreatment algorithm was presented for the lower accuracy problem compared with the original depth information in the field of three-Dimensional (3D) scene measurement for robot's perception. Firstly, a measuring and sampling model of the depth map was developed to realize the Monte Carlo uncertainty evaluation model. Secondly, the depth value intervals were calculated to judge and filter the noise pixels. Finally, noise points were repaired with mean-value of the estimation intervals. The experimental results show that the algorithm can effectively suppress and repair the noise pixels while keeping the depth gradient and values of non-noise pixels. The Mean Square Error (MSE) of depth map after preprocessing is reduced by 15.25% to 28.79%, and the object profiles remain unchanged compared with the JBF (Joint Bilateral Filtering) based on color and depth map. Therefore, it achieves the purpose of improving the depth information accuracy in 3D scenes.

Key words: depth map, Monte Carlo model, uncertainty evaluation, noise filtering, depth recovery

中图分类号: