Search

Accurate tree species identification is essential for ecological evaluation and other forest applications. In this paper, we proposed a point-based deep neural network called LayerNet. For light detection and ranging (LiDAR) data in forest regions, the network can divide multiple overlapping layers in Euclidean space to obtain the local three-dimensional (3D) structural features of the tree. The features of all layers are aggregated, and the global feature is obtained by convolution to classify the tree species. To validate the proposed framework, multiple experiments, including airborne and ground-based LiDAR datasets, are conducted and compared with several existing tree species classification algorithms. The test results show that LayerNet can directly use 3D data to accurately classify tree species, with the highest classification accuracy of 92.5%. Also, the results of comparative experiments demonstrate that the proposed framework has obvious advantages in classification accuracy and provides an effective solution for tree species classification tasks.

Remote sensing (RS) image segmentation is an essential step in geographic object-based image analysis (GEOBIA) to ultimately derive “meaningful objects”. While many segmentation methods exist, most of them are not efficient for large data sets. Thus, the goal of this research is to develop an efficient parallel multi-scale segmentation method for RS imagery by combining graph theory and the fractal net evolution approach (FNEA). Specifically, a minimum spanning tree (MST) algorithm in graph theory is proposed to be combined with a minimum heterogeneity rule (MHR) algorithm that is used in FNEA. The MST algorithm is used for the initial segmentation while the MHR algorithm is used for object merging. An efficient implementation of the segmentation strategy is presented using data partition and the “reverse searching-forward processing” chain based on message passing interface (MPI) parallel technology. Segmentation results of the proposed method using images from multiple sensors (airborne, SPECIM AISA EAGLE II, WorldView-2, RADARSAT-2) and different selected landscapes (residential/industrial, residential/agriculture) covering four test sites indicated its efficiency in accuracy and speed. We conclude that the proposed method is applicable and efficient for the segmentation of a variety of RS imagery (airborne optical, satellite optical, SAR, high-spectral), while the accuracy is comparable with that of the FNEA method. (VLID)2600370