Automated analysis of in situ canopy images for the estimation of forest canopy cover

Abstract

Forest canopy cover is a commonly used ecological indicator variable, but its reliable estimation in the field with, e.g., the line intersect sampling method is laborious. One alternative for obtaining this information quickly is to sample a larger area of canopy from each prespecified location within a sample plot by using digital cameras and then use image analysis techniques to extract canopy cover from the images. The first aim of this study was to implement an automated method for the estimation of canopy cover from skyward-looking canopy images obtained with standard digital cameras. The second aim was to compare the estimates obtained with the cameras and the automated analysis technique to control values measured with a vertical tube by using two different plot types and a varying angular field of view (1-50°). The results indicate that the automated analysis technique, involving an automated image thresholding algorithm and morphological image processing operations to eliminate the effect of within-crown gaps, can reliably replace manual postprocessing. With the tested sampling schemes, a 40° angle of view yielded the best approximation of the control. © 2009 by the Society of American Foresters.