Laser Range-, Color-, and Texture-based Classifiers for Segmenting Marginal Roads

Published

December 7, 2001

Author(s)

C E. Rasmussen

Abstract

We describe preliminary results on combining depth information from a laser range-finder and color and texture image cues to train classifiers to segment ill-structured dirt, gravel, and asphalt roads as input to an autonomous road following system. A large number of registered laser and camera images were captured at frame-rate on a variety of rural roads, allowing laser features such as 3-D height and smoothness to be correlated with image features such color histograms and Gabor filter responses. A small set of road models were generated by training separate neural networks on labeled feature vectors clustered by road ``type.'' By first classifying the type of a novel road image, an appropriate second-stage classifier was selected to segment individual pixels, achieving a high degree of accuracy on arbitrary images from the dataset.

Proceedings Title

Conference on Computer Vision and Pattern Recognition Technical Sketches

Conference Dates

December 1, 2001

Conference Location

Kauai, HI

Conference Title

Computer Vision and Pattern Recognition Technical Sketches

Pub Type

Conferences

Download Paper

Local Download

Keywords

3D modeling, 3D modeling, 3D modeling, 3D modeling, 3D modeling, Aerial Work Platforms, Aerial Work Platforms, artificial intelligence, Autonomous mobility, Autonomous mobility, Autonomous mobility, Autonomous mobility, Autonomous mobility, Boomlifts, Boomlifts, complexity, dimensional, dimensional, dimensional, dimensional, dimensional, dimensional, dimensional, dimensional, dimensional, dimensional, dimensional, Distributed Motion Control, Distributed Motion Control, DMIS, DMIS, DMIS, DMIS, DMIS, DMIS, DMIS, DMIS, DMIS, DMIS, DMIS, emerging behaviors, Inexpensive Servoes, Inexpensive Servoes, inspection, inspection, inspection, inspection, inspection, inspection, inspection, inspection, inspection, inspection, inspection, Intell, intelligent control, intelligent control, intelligent control, intelligent control, intelligent control, Intelligent Systems, Large Manipulators, Large Manipulators, machine intelligence, metrology, metrology, metrology, metrology, metrology, metrology, metrology, metrology, metrology, metrology, metrology, mobile robots, mobile robots, mobile robots, mobile robots, mobile robots, multiresolutional hierarchical control, NIST, NIST, NIST, NIST, NIST, NIST, NIST, NIST, NIST, NIST, NIST, numerical control, numerical control, numerical control, numerical control, numerical control, numerical control, numerical control, numerical control, numerical control, numerical control, numerical control, ontologies, Paint Stripping, Paint Stripping, perception, perception, perception, perception, perception, Performance Metrics, Planning, process-planning, process-planning, process-planning, process-planning, process-planning, process-planning, process-planning, process-planning, process-planning, process-planning, process-planning, RCS, sensors, sensors, sensors, sensors, sensors, standard, standard, standard, standard, standard, standard, standard, standard, standard, standard, standard, world model, world model, world model, world model, world model

Citation

Rasmussen, C. (2001), Laser Range-, Color-, and Texture-based Classifiers for Segmenting Marginal Roads, Conference on Computer Vision and Pattern Recognition Technical Sketches, Kauai, HI, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=821649 (Accessed July 27, 2024)