<p><strong>Abstract.</strong> Most methods for the mapping of tree species are based on the segmentation of single trees that are subsequently classified using a set of hand-crafted features and an appropriate classifier. The classification accuracy for coniferous and deciduous trees just using airborne laser scanning (ALS) data is only around 90% in case the geometric information of the point cloud is used. As deep neural networks (DNNs) have the ability to adaptively learn features from the underlying data, they have outperformed classic machine learning (ML) approaches on well-known benchmark datasets provided by the robotics, computer vision and remote sensing community. Though, tree species classification using deep learning (DL) procedures has been of minor research interest so far. Some studies have been conducted based on an extensive prior generation of images or voxels from the 3D raw data. Since innovative DNNs directly operate on irregular and unordered 3D point clouds on a large scale, the objective of this study is to exemplarily use PointNet++ for the semantic labeling of ALS point clouds to map deciduous and coniferous trees. The dataset for our experiments consists of ALS data from the Bavarian Forest National Park (366 trees/ha), only including spruces (coniferous) and beeches (deciduous). First, the training data were generated automatically using a classic feature-based Random Forest (RF) approach classifying coniferous trees (precision&thinsp;=&thinsp;93%, recall&thinsp;=&thinsp;80%) and deciduous trees (precision&thinsp;=&thinsp;82%, recall&thinsp;=&thinsp;92%). Second, PointNet++ was trained and subsequently evaluated using 80 randomly chosen test batches à 400&thinsp;m<sup>2</sup>. The achieved per-point classification results after 163 training epochs for coniferous trees (precision&thinsp;=&thinsp;90%, recall&thinsp;=&thinsp;79%) and deciduous trees (precision&thinsp;=&thinsp;81%, recall&thinsp;=&thinsp;91%) are fairly high considering that only the geometry was included. Nevertheless, the classification results using PointNet++ are slightly lower than those of the baseline method using a RF classifier. Errors in the training data and occurring edge effects limited a better performance. Our first results demonstrate that the architecture of the 3D DNN PointNet++ can successfully be adapted to the semantic labeling of large ALS point clouds to map deciduous and coniferous trees. Future work will focus on the integration of additional features like i.e. the laser intensity, the surface normals and multispectral features into the DNN. Thus, a further improvement of the accuracy of the proposed approach is to be expected. Furthermore, the classification of numerous individual tree species based on pre-segmented single trees should be investigated.</p>
Automated Visual Recognizability Evaluation of Traffic Sign Based on 3D LiDAR Point Clouds
