ABSTRACTThe distribution of leafless and full-canopy trees in fall is essential in understanding the phenological and morphological characteristics of forests and those are critical baseline information for several scientific questions and management scenarios. The rapid progress of very high-resolution satellite-borne sensors provides a feasible solution to fast and accurately map tree coverage during the growing season. However, less attention has been paid to the leaf-off seasons. One biggest challenge with classification on fall imagery is the potential confusion among different land features, as evidenced by the no significant difference between bare land and leafless trees from t-test and transformed divergence (TD) analysis results. We proposed an integrative approach to classifying leafless and full-canopy trees on fall imagery of 0.46-m WorldView-2 (WV-2). We adopted a two-step classification approach by first removing the non-forest area from the WV-2 imagery with the mask layer generated from the publicly available summer imagery of 1-m National Agriculture Imagery Program imagery via object-based image analysis. Two classification methods, namely decision tree (DT) and nearest neighborhood (NN), were employed to classify full canopy and leafless trees in the masked WV-2 imagery. DT performed slightly better than NN regarding higher overall accuracy and fewer feature requirements.
Read full abstract