Relationship between dbh and crown characteristics derived by airborne laser scanner

The main goal of the study was the determination of the relationship between selected crown parameters derived from automatically processed airborne laser scanning (ALS) data and tree diameter. It has been found that the best set of features for that purpose is crown projection area (CPA) as a predictor and tree basal area (BA) as dependent variable. Two algorithms of crown segmentation (‘growing’ and ‘pouring’) have been analysed. Weibull, exponential and hyperbole functions have been the best for description of the relationship between CPA and BA. The relationship between tree diameter and crown characteristics was significant, but relatively weak. This can be attributed to the fact, that the relationship can be modified by various stand characteristics. Including such variables as tree stocking and crown closure should improve the model significantly