Semi-automated Forest Stand Delineation Usingwavelet-based Segmentation of Very High Resolution Optical Imagery in Flanders, Belgium

Stand delineation estimation is one of the cornerstones of forest inventory mapping and a key element to forest management decision making. Stands are forest management units defined mainly by similar species composition, density, closure, height and age. Stand boundaries (and also attributes) are traditionally estimated through air photo interpretation. Visual interpretation is intelligibly subjective and can be remedied by numerical interpretation through automated image processing. In this paper, a stand delineation method is presented integrating wavelet analysis into the image segmentation process. It is novel in the sense that no direct spectral information is included. Using both wavelet coefficients and derived statistics, like mean absolute value and standard deviation, allowed for discrimination between forest compartments that differ in the above mentioned attributes. This approach was developed using simulated forest stands and was subsequently applied to digital aerial photographs of a forest site (representing a mixture of softand hardwood stands) in Flanders, Belgium. The presented method was qualitatively evaluated against traditional image segmentation i.e. segmentation based on the images’ spectral information. It is concluded that the proposed method outperformed traditional image segmentation. In addition, this research was valuable to assess the added value of wavelet coefficients in object-based image segmentation.