Classification and Segmentation of Mining Area Objects in Large-Scale Spares Lidar Point Cloud Using a Novel Rotated Density Network

Deep Neural Network-Based Phase-Modulated Continuous-Wave LiDAR

LiDAR has high accuracy and resolution and is widely used in various fields. In particular, phase-modulated continuous-wave (PhMCW) LiDAR has merits such as low power, high precision, and no need for laser frequency modulation. However, with decreasing signal-to-noise ratio (SNR), the noise on the signal waveform becomes so severe that the current methods to extract the time-of-flight are no longer feasible. In this paper, a novel method that uses deep neural networks to measure the pulse width is proposed. The effects of distance resolution and SNR on the performance are explored. Recognition accuracy reaches 81.4% at a 0.1 m distance resolution and the SNR is as low as 2. We simulate a scene that contains a vehicle, a tree, a house, and a background located up to 6 m away. The reconstructed point cloud has good fidelity, the object contours are clear, and the features are restored. More precisely, the three distances are 4.73 cm, 6.00 cm, and 7.19 cm, respectively, showing that the performance of the proposed method is excellent. To the best of our knowledge, this is the first work that employs a neural network to directly process LiDAR signals and to extract their time-of-flight.

A Classification Method of Point Clouds of Transmission Line Corridor Based on Improved Random Forest and Multi-Scale Features

Classification of airborne laser scanning (ALS) point clouds of power lines is of great importance to their reconstruction. However, it is still a difficult task to efficiently and accurately classify the ground, vegetation, power lines and power pylons from ALS point clouds. Therefore, in this paper, a method is proposed to improve the accuracy and efficiency of the classification of point clouds of transmission lines, which is based on improved Random Forest and multi-scale features. The point clouds are filtered by the optimized progressive TIN densification filtering algorithm, then the elevations of the filtered point cloud are normalized. The features of the point cloud at different scales are calculated according to the basic features of the point cloud and the characteristics of transmission lines. The Relief F and Sequential Backward Selection algorithm are used to select the best subset of features to estimate the parameters of the learning model, then an Improved Random Forest classification model is built to classify the point clouds. The proposed method is verified by using three different samples from the study area and the results show that, compared with the methods based on Support Vector Machines, AdaBoost or Random Forest, our method can reduce feature redundancy and has higher classification accuracy and efficiency.

An Efficient and General Framework for Aerial Point Cloud Classification in Urban Scenarios

With recent advances in technologies, deep learning is being applied more and more to different tasks. In particular, point cloud processing and classification have been studied for a while now, with various methods developed. Some of the available classification approaches are based on specific data source, like LiDAR, while others are focused on specific scenarios, like indoor. A general major issue is the computational efficiency (in terms of power consumption, memory requirement, and training/inference time). In this study, we propose an efficient framework (named TONIC) that can work with any kind of aerial data source (LiDAR or photogrammetry) and does not require high computational power while achieving accuracy on par with the current state of the art methods. We also test our framework for its generalization ability, showing capabilities to learn from one dataset and predict on unseen aerial scenarios.