Measuring vegetation along railway tracks

This paper presents a novel system for measuring vegetation along railway tracks. The system, which is mounted on a train, consists of three multi-spectral cameras that are sensitive in the visible as well as in the near-infrared spectrum. The goal is to generate a vegetation profile and vegetation register for the efficient deployment of staff to eliminate vegetation. For this, two different problems are solved: The distinction of vegetation and non-vegetation and the SB-measurement of the vegetation. The use of multi-spectral cameras is necessary for the distinction of green vegetation and the green masts of the overhead contact system. The system segments vegetation correctly in 86% of the vegetated image regions. Only 0.4% of non-vegetation pixels are falsely classified. SD-measurement is performed by triangulation.

[1]  Michel Humbert,et al.  Optimization of the Positivity Method in Quantitative Texture Analysis , 1990 .

[2]  Anil K. Jain,et al.  Texture Analysis , 2018, Handbook of Image Processing and Computer Vision.

[3]  Alonzo Kelly,et al.  Terrain typing for real robots , 1995, Proceedings of the Intelligent Vehicles '95. Symposium.

[4]  Pierre Charbonnier,et al.  Road boundaries detection using color saturation , 1998, 9th European Signal Processing Conference (EUSIPCO 1998).

[5]  Anil K. Jain,et al.  Detecting sky and vegetation in outdoor images , 1999, Electronic Imaging.

[6]  D. Roberts,et al.  Integrating polarimetric synthetic aperture radar and imaging spectrometry for wildland fuel mapping in southern California , 2000 .

[7]  K. Wroz,et al.  Erfahrungen mit dem LIMEZ II: Der Lichtraummesszug der Deutschen Bahn im Einsatz , 2000 .

[8]  John A. Marchant,et al.  Physics-based colour image segmentation for scenes containing vegetation and soil , 2001, Image Vis. Comput..

[9]  Martial Hebert,et al.  Terrain Classification Techniques From Ladar Data For Autonomous Navigation , 2002 .

[10]  Alain Clément,et al.  Unsupervised segmentation of scenes containing vegetation (Forsythia) and soil by hierarchical analysis of bi-dimensional histograms , 2003, Pattern Recognit. Lett..

[11]  B. Hulin,et al.  Improvements on obstacle detection in the pantograph gauge due to the recognition of steady arms , 2004, IEEE Intelligent Vehicles Symposium, 2004.

[12]  D. Roberts,et al.  Spectral and Structural Measures of Northwest Forest Vegetation at Leaf to Landscape Scales , 2004, Ecosystems.

[13]  V. Appelt,et al.  Berechnung von Signalsichtweiten: Neue Ansätze auf der Grundlage von 3D-Modellen , 2004 .