Metadata only
Datum
2007Typ
- Conference Paper
ETH Bibliographie
yes
Altmetrics
Abstract
Forest boundary delineation is one of the key issues of forest management for Swiss National Forest Inventory (NFI). The proposed approach in this paper focuses on the detection of forest boundaries with special emphasis on spatially contiguous and reproducible results by using both aerial images and LIDAR data. The used Green Vegetation Index (GVI) is helpful to find green vegetation areas while the curvature feature curvature features extracted from the canopy height model (CHM) is helpful for building removing, The combination of curvature features, GVI, JSEG segmentation result and the Gabor wavelet texture features leads to forest area detection and obtains proper forest boundaries. Preliminary results are encouraging, regarding an automatic process and delivers robust forest boundary detection. Mehr anzeigen
Publikationsstatus
publishedBuchtitel
Lecture Notes in Signal Science, Internet and EducationZeitschrift / Serie
Proceedings of the WSEAS International ConferencesSeiten / Artikelnummer
Verlag
World Scientific and Engineering Academy and SocietyKonferenz
Thema
Aerial images; Image segmentation; LIDAR; Gabor wavelet; Forest inventory; Texture featuresETH Bibliographie
yes
Altmetrics