Multi-sensor calibration through iterative registration and fusion

Authors:
Yunbao Huang;Xiaoping Qian;Shiliang Chen
Affiliations:
Mechanical, Materials & Aerospace Engineering Department, Illinois Institute of Technology, United States and CAD Center of Huazhong University of Science & Technology, Wuhan, China;Mechanical, Materials & Aerospace Engineering Department, Illinois Institute of Technology, United States;Mechanical, Materials & Aerospace Engineering Department, Illinois Institute of Technology, United States
Venue:
Computer-Aided Design
Year:
2009

Citing 22
Cited 1

A new algorithm for spline smoothing based on smoothing a stochastic process

SIAM Journal on Scientific and Statistical Computing - Papers from the Second Conference on Parallel Processing for Scientific Computin
A Method for Registration of 3-D Shapes

IEEE Transactions on Pattern Analysis and Machine Intelligence - Special issue on interpretation of 3-D scenes—part II
On 3D model construction by fusing heterogeneous sensor data

CVGIP: Image Understanding
Iterative point matching for registration of free-form curves and surfaces

International Journal of Computer Vision
The NURBS book

The NURBS book
Towards a General Multi-View Registration Technique

IEEE Transactions on Pattern Analysis and Machine Intelligence
Simultaneous registration of multiple range views for use in reverse engineering of CAD models

Computer Vision and Image Understanding - Special issue on CAD-based computer vision
ICP Registration Using Invariant Features

IEEE Transactions on Pattern Analysis and Machine Intelligence
Real-time 3D model acquisition

Proceedings of the 29th annual conference on Computer graphics and interactive techniques
A method for fine registration of multiple view range images considering the measurement error properties

Computer Vision and Image Understanding - Registration and fusion of range images
An Introduction to the Kalman Filter

An Introduction to the Kalman Filter
Geometry and Convergence Analysis of Algorithms for Registration of 3D Shapes

International Journal of Computer Vision
Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning)

Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning)
Registration of Multiple Point Sets

ICPR '96 Proceedings of the 13th International Conference on Pattern Recognition - Volume 2
Total least squares fitting of point sets in m-D

CGI '05 Proceedings of the Computer Graphics International 2005
Global registration of multiple 3D point sets via optimization-on-a-manifold

SGP '05 Proceedings of the third Eurographics symposium on Geometry processing
Robust global registration

SGP '05 Proceedings of the third Eurographics symposium on Geometry processing
Bayesian surface reconstruction via iterative scan alignment to an optimized prototype

SGP '07 Proceedings of the fifth Eurographics symposium on Geometry processing
Dynamic B-spline surface reconstruction: Closing the sensing-and-modeling loop in 3D digitization

Computer-Aided Design
Multiview registration for large data sets

3DIM'99 Proceedings of the 2nd international conference on 3-D digital imaging and modeling
Least-squares estimation: from Gauss to Kalman

IEEE Spectrum
Multiview registration of 3D scenes by minimizing error between coordinate frames

IEEE Transactions on Pattern Analysis and Machine Intelligence

Point set augmentation through fitting for enhanced ICP registration of point clouds in multisensor coordinate metrology

Robotics and Computer-Integrated Manufacturing

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a new multi-sensor calibration approach, called iterative registration and fusion (IRF), is presented. The key idea of this approach is to use surfaces reconstructed from multiple point clouds to enhance the registration accuracy and robustness. It calibrates the relative position and orientation of the spatial coordinate systems among multiple sensors by iteratively registering the discrete 3D sensor data against an evolving reconstructed B-spline surface, which results from the Kalman filter-based multi-sensor data fusion. Upon each registration, the sensor data gets closer to the surface. Upon fusing the newly registered sensor data with the surface, the updated surface represents the sensor data more accurately. We prove that such an iterative registration and fusion process is guaranteed to converge. We further demonstrate in experiments that the IRF can result in more accurate and more stable calibration than many classical point cloud registration methods.