A decision network based frame-work for visual off-road path detection problem
- 1 January 2006
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
Abstract
This paper describes a decision network based frame-work used for path-detection algorithm development in autonomous vehicle applications. Lane marker detection algorithms do not work in off-road environments. Off-road trails have too much complexity, with widely varying textures and many differing natural boundaries. The authors have developed a general approach. Images are segmented into regions, based on the homogeneity of some pixel properties and the resulting regions are classified as road or not-road by a decision network process. Combinations of contiguous clusters form the path surface, allowing any arbitrary path to be representedKeywords
This publication has 7 references indexed in Scilit:
- An agent based evolutionary approach to path detection for off-road vehicle guidancePattern Recognition Letters, 2006
- Soft Learning Vector Quantization and Clustering Algorithms Based on Non-Euclidean Norms: Single-Norm AlgorithmsIEEE Transactions on Neural Networks, 2005
- Locally linear discriminant analysis for multimodally distributed classes for face recognition with a single model imagePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2005
- Obstacle detection using adaptive color segmentation and color stereo homographyPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- A fast and robust vision based road following algorithmPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- GOLD: a parallel real-time stereo vision system for generic obstacle and lane detectionIEEE Transactions on Image Processing, 1998
- Decision Analytic Networks in Artificial IntelligenceManagement Science, 1995