Model-based conifer canopy surface reconstruction from photographic imagery: Overcoming the occlusion, foreshortening, and edge effects

Abstract

Canopy surface data are desirable in forestry, but they are difficult to collect in the field. Existing surface reconstruction techniques cannot adequately extract canopy surfaces, especially for conifer stands. This paper develops an integrated model-based approach to reconstruct canopy surface for conifer stands analytically from the crown level. To deal with dense stands, critical problems are addressed in the process of model-based surface reconstruction. These include the occlusion problem in disparity (parallax) prediction from tree models, the edge effect of tree models on the disparity map, and the foreshortening effect in image matching. The model-based approach was applied to recover the canopy surface of a dense redwood stand using images scanned from 1:2,400-scale aerial photographs. Compared with field measurements, crown radius and tree height derived from the reconstructed canopy surface model have an overall accuracy of 92 percent and 94 percent, respectively. The results demonstrate the approach's ability to reconstruct complicated stands.