Remote sensing as a tool for monitoring plant invasions: testing the effects of data resolution and image classification approach on the detection of a model plant species Heracleum mantegazzianum (giant hogweed)

0397795 - BÚ 2014 RIV NL eng J - Journal Article
Müllerová, Jana - Pergl, Jan - Pyšek, Petr
Remote sensing as a tool for monitoring plant invasions: testing the effects of data resolution and image classification approach on the detection of a model plant species Heracleum mantegazzianum (giant hogweed).
International Journal of Applied Earth Observation and Geoinformation. Roč. 25, Dec.2013 (2013), s. 55-65. ISSN 0303-2434. E-ISSN 1872-826X
R&D Projects: GA AV ČR IAA600050811
Institutional support: RVO:67985939
Keywords : historical aerial VHR photography * invasion progress * object and pixel-based image classification
Subject RIV: EF - Botanics
Impact factor: 2.539, year: 2013

Remote sensing assessment of a noxious invasive species Heracleum mantegazzianum integrating different data sources, spatial and spectral resolutions, and image processing techniques. Panchromatic, multispectral and color very high spatial resolution aerial photography (1947 to 2006, res. 0.5 m), and medium spatial resolution satellite data (Rapid Eye 2010, res. 5 m) were analyzed to assess their potential for hogweed monitoring by using pixel- and object-based image analysis. We established automatic, relatively fast, and efficient (time-effective and standardized) computer-assisted methods for monitoring invasions and their long-term dynamics that could be applied over large areas. The methodology presented here is suitable for mapping the progress of invasion, identifying areas at the highest invasion risk, and enabling early detection at the beginning of invasion, hence it can make management measures more efficient and less expensive.
Permanent Link: http://hdl.handle.net/11104/0225992