30 April 1992 Three-dimensional range image segmentation and fitting by quadratic surfaces
Author Affiliations +
Proceedings Volume 1611, Sensor Fusion IV: Control Paradigms and Data Structures; (1992); doi: 10.1117/12.57954
Event: Robotics '91, 1991, Boston, MA, United States
Abstract
This paper presents a robust segmentation and fitting technique. The method randomly samples appropriate range image points and fits them into selected primitive type. From K samples we measure residual consensus to choose one set of sample points which determines an equation to have the best fit for a homogeneous patch in the current processing region. A method with compressed histogram is used to measure and compare residuals on various noise levels. The method segments range image into quadratic surfaces, and works very well even in smoothly connected regions.
© (1992) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xinming Yu, Tien D. Bui, Adam Krzyzak, "Three-dimensional range image segmentation and fitting by quadratic surfaces", Proc. SPIE 1611, Sensor Fusion IV: Control Paradigms and Data Structures, (30 April 1992); doi: 10.1117/12.57954; https://doi.org/10.1117/12.57954
PROCEEDINGS
10 PAGES


SHARE
KEYWORDS
Image segmentation

3D image processing

Sensor fusion

Image fusion

Sensors

Image processing

Image processing algorithms and systems

RELATED CONTENT

Multiresolution range-guided stereo matching
Proceedings of SPIE (April 01 1991)
Vision system for ALV
Proceedings of SPIE (September 25 1998)
Joint high dynamic range imaging and color demosaicing
Proceedings of SPIE (October 26 2011)
Image segmentation based on the fusion technique
Proceedings of SPIE (July 31 2002)

Back to Top