Point Clouds Classification In Road Environment Based On Deep Learning

The automatic classification of 3D point clouds is publicly known as a challenging task in a complex road environment.Specifically,each point is automatically classified into a unique category label,and then,the labels are used as clues for semantic analysis and scene recognition.Instead of heuristically extracting handcrafted features in traditional methods to classify all points,we put forward an end-to-end octree-based fully convolutional network(FCN)to classify 3D point clouds in an urban road environment.There are four contributions of our work.The first is that the integration and comprehensive uses of OctNet and FCN greatly decrease the computing time and memory demands compared with a dense 3D convolutional neural network(CNN).The second is that the octree-based network is strengthened by means of modifying the cross-entropy loss function to solve the problems of an unbalanced category distribution.The third is that an Inception-ResNet block is united with our network,which enables our 3D CNN to effectively learn how to classify scenes containing objects at multiple scales and improve classification accuracy.The last is that an open source dataset(HuangshiRoad dataset)with 10 different classes is introduced for 3D point cloud classification.Three representative datasets(Semantic3D,WHU_MLS(block I and block II)and HuangshiRoad)with different covered areas and numbers of points and classes are selected to evaluate our proposed method.The experimental results show that the overall classification accuracy is appreciable,with 89.4% for Semantic3 D,82.9% for WHU_MLS block I,91.4% for WHU_MLS block II and 94% for HuangshiRoad.Our deep learning approach can efficiently classify 3D dense point clouds in an urban road environment measured by a mobile laser scanning(MLS)system or static LiDAR.