Object-Oriented Classification of Forest Structure from Light Detection and Ranging Data for Stand Mapping

Authors: Sullivan, Alicia A.; McGaughey, Robert J.; Andersen, Hans-Erik; Schiess, Peter

Source: Western Journal of Applied Forestry, Volume 24, Number 4, October 2009 , pp. 198-204(7)

Publisher: Society of American Foresters

Buy & download fulltext article:

Price: $29.50 plus tax (Refund Policy)

Abstract:

Stand delineation is an important step in the process of establishing a forest inventory and provides the spatial framework for many forest management decisions. Many methods for extracting forest structure characteristics for stand delineation and other purposes have been researched in the past, primarily focusing on high-resolution imagery and satellite data. High-resolution airborne laser scanning offers new opportunities for evaluating forests and conducting forest inventory. This study investigates the use of information derived from light detection and ranging (LIDAR) data as a potential tool for delineation of forest structure to create stand maps. Delineation methods are developed and tested using data sets collected over the Blue Ridge study site near Olympia, Washington. The methodology developed delineates forest areas using LIDAR data and object-oriented image segmentation and supervised classification. Error matrices indicate classification accuracies with a kappa hat values of 78 and 84% for 1999 and 2003 data sets, respectively.

Keywords: LIDAR; light detection and ranging; forest classification; object-oriented image segmentation; supervised classification; stand mapping

Document Type: Research article

Publication date: 2009-10-01

More about this publication?