Peng Ge
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Peng Ge has a total of 11 co-authors. Here are the top co-authors by number of shared publications:
- Bing Han
Sino-European Institute of Aviation Engineering, Civil Aviation University of China, Tianjin 300300, China
4 shared publications - Jingtang Hao
Sino-European Institute of Aviation Engineering, Civil Aviation University of China, Tianjin 300300, China
3 shared publications - Zhigang Su
Sino-European Institute of Aviation Engineering, Civil Aviation University of China, Tianjin 300300, China
3 shared publications - Yue Wang
Information Countermeasure Technology Laboratory, Beijing Research Institute of Telemetry, Beijing 100076, China
2 shared publications - Chengxu Hu
Sino-European Institute of Aviation Engineering, Civil Aviation University of China, Tianjin 300300, China
1 shared publication - Hongchang Wang
Sino-European Institute of Aviation Engineering, The Civil Aviation University of China, Tianjin 300300, China
1 shared publication
Latest Publications
- Open Access
NELD-EC: Neighborhood-Effective-Line-Density-Based Euclidean Clustering for Point Cloud Segmentation
For the problem that it is difficult to effectively cluster lidar point clouds with irregular shapes and uneven densities, a Neighborhood Effective Line Density (NELD)-based Euclidean Clustering (NELD-EC) algorithm is proposed in this paper. The NELD-EC algorithm first eliminates the interfering points within the neighborhood of the data point by utilizing the distance relationship and calculates the NELD of the data point using the effective neighborhood set without interfering points of the data point. The NELD of a data point is taken as the local density of that data point. Then, the NELD-EC algorithm conducts clustering processing using the NELD of all data points and uses the reciprocal of the harmonic average of the local densities of all data points within each cluster after clustering as the distance threshold for the data points within the cluster. Finally, the NELD-EC algorithm completes the clustering of the point cloud based on the adjusted adaptive distance threshold. The clustering experimental results on simulated point clouds, fixed point clouds, and sequential point clouds indicate that, compared with several other typical Euclidean clustering algorithms, the NELD-EC algorithm requires simpler parameters to be set, is less sensitive to the initial distance threshold, can effectively reduce the occurrence probabilities of over-segmentation and under-segmentation, and has strong stability in clustering performance. The NELD-EC algorithm is more suitable for processing sequential point clouds in actual dynamic and complex scenarios.
Open AccessNELD-EC: Neighborhood-Effective-Line-Density-Based Euclidean Clustering for Point Cloud Segmentation
For the problem that it is difficult to effectively cluster lidar point clouds with irregular shapes and uneven densities, a Neighborhood Effective Line Density (NELD)-based Euclidean Clustering (NELD-EC) algorithm is proposed in this paper. The NELD-EC algorithm first eliminates the interfering points within the neighborhood of the data point by utilizing the distance relationship and calculates the NELD of the data point using the effective neighborhood set without interfering points of the data point. The NELD of a data point is taken as the local density of that data point. Then, the NELD-EC algorithm conducts clustering processing using the NELD of all data points and uses the reciprocal of the harmonic average of the local densities of all data points within each cluster after clustering as the distance threshold for the data points within the cluster. Finally, the NELD-EC algorithm completes the clustering of the point cloud based on the adjusted adaptive distance threshold. The clustering experimental results on simulated point clouds, fixed point clouds, and sequential point clouds indicate that, compared with several other typical Euclidean clustering algorithms, the NELD-EC algorithm requires simpler parameters to be set, is less sensitive to the initial distance threshold, can effectively reduce the occurrence probabilities of over-segmentation and under-segmentation, and has strong stability in clustering performance. The NELD-EC algorithm is more suitable for processing sequential point clouds in actual dynamic and complex scenarios.
- Open Access
Research on Ground Object Echo Simulation of Avian Lidar
The clutter suppression effect of ground objects significantly impacts the detection and tracking performance of avian lidar on low-altitude bird flock targets. It is imperative to simulate the point cloud data of ground objects in lidar to explore effective methods for suppressing clutter caused by ground objects in avian lidar. The traditional ray-tracing method is enhanced in this paper to efficiently obtain the point cloud simulation results of ground objects. By incorporating a beam constraint and a light-energy constraint, the screening efficiency of effective rays is improved, making them more suitable for simulating large scenes with narrow lidar beams. In this paper, a collision detection scheme is proposed based on beam constraints, aiming to significantly enhance the efficiency of ray-tracing collision detection. The simulation and experimental results demonstrate that, in comparison with other conventional simulation methods, the proposed method yields the point cloud results of ground objects that exhibit greater conformity to the actual lidar-collected point cloud results in terms of shape characteristics and intensity features. Additionally, the simulation speed is significantly enhanced.
Open AccessResearch on Ground Object Echo Simulation of Avian Lidar
The clutter suppression effect of ground objects significantly impacts the detection and tracking performance of avian lidar on low-altitude bird flock targets. It is imperative to simulate the point cloud data of ground objects in lidar to explore effective methods for suppressing clutter caused by ground objects in avian lidar. The traditional ray-tracing method is enhanced in this paper to efficiently obtain the point cloud simulation results of ground objects. By incorporating a beam constraint and a light-energy constraint, the screening efficiency of effective rays is improved, making them more suitable for simulating large scenes with narrow lidar beams. In this paper, a collision detection scheme is proposed based on beam constraints, aiming to significantly enhance the efficiency of ray-tracing collision detection. The simulation and experimental results demonstrate that, in comparison with other conventional simulation methods, the proposed method yields the point cloud results of ground objects that exhibit greater conformity to the actual lidar-collected point cloud results in terms of shape characteristics and intensity features. Additionally, the simulation speed is significantly enhanced.
- Open Access
Target Detection in Single-Photon Lidar Using CNN Based on Point Cloud Method
To enhance the detection capability of weak targets and reduce the dependence of single-photon lidar target detection on the number of the time-correlated single-photon counting detection cycles, a convolutional neural network (CNN) based on the point cloud (CNN-PC) method is proposed in this paper for detecting targets in single-photon lidar. This approach utilizes the exceptional feature extraction capabilities offered by CNN. The CNN-PC method utilizes the feature extraction module of the trained CNN to simultaneously extract features from two-dimensional point cloud slices. Subsequently, it combines these features and feeds them into the classification module of the trained CNN for final target detection. By training the CNN using point cloud slices generated with a minimal number of detection cycles and employing a parallel structure to extract features from multiple point cloud slices, the CNN-PC method exhibits remarkable flexibility in adapting to varying numbers of detection cycles. Both simulation and experimental results demonstrate that the CNN-PC method outperforms the classical constant false alarm rate method in terms of the target detection probability at the same signal-to-noise ratio and in terms of the imaging rate and error rate at the same number of detection cycles.
Open AccessTarget Detection in Single-Photon Lidar Using CNN Based on Point Cloud Method
To enhance the detection capability of weak targets and reduce the dependence of single-photon lidar target detection on the number of the time-correlated single-photon counting detection cycles, a convolutional neural network (CNN) based on the point cloud (CNN-PC) method is proposed in this paper for detecting targets in single-photon lidar. This approach utilizes the exceptional feature extraction capabilities offered by CNN. The CNN-PC method utilizes the feature extraction module of the trained CNN to simultaneously extract features from two-dimensional point cloud slices. Subsequently, it combines these features and feeds them into the classification module of the trained CNN for final target detection. By training the CNN using point cloud slices generated with a minimal number of detection cycles and employing a parallel structure to extract features from multiple point cloud slices, the CNN-PC method exhibits remarkable flexibility in adapting to varying numbers of detection cycles. Both simulation and experimental results demonstrate that the CNN-PC method outperforms the classical constant false alarm rate method in terms of the target detection probability at the same signal-to-noise ratio and in terms of the imaging rate and error rate at the same number of detection cycles.
- Open Access
A Gated-Recurrent-Unit-Based Interacting Multiple Model Method for Small Bird Tracking on Lidar System
Lidar presents a promising solution for bird surveillance in airport environments. However, the low observation refresh rate of Lidar poses challenges for tracking bird targets. To address this problem, we propose a gated recurrent unit (GRU)-based interacting multiple model (IMM) approach for tracking bird targets at low sampling frequencies. The proposed method constructs various GRU-based motion models to extract different motion patterns and to give different predictions of target trajectory in place of traditional target moving models and uses an interacting multiple model mechanism to dynamically select the most suitable GRU-based motion model for trajectory prediction and tracking. In order to fuse the GRU-based motion model and IMM, the approximation state transfer matrix method is proposed to transform the prediction of GRU-based network into an explicit state transfer model, which enables the calculation of the models’ probability. The simulation carried out on an open bird trajectory dataset proves that our method outperforms classical tracking methods at low refresh rates with at least 26% improvement in tracking error. The results show that the proposed method is effective for tracking small bird targets based on Lidar systems, as well as for other low-refresh-rate tracking systems.
Open AccessA Gated-Recurrent-Unit-Based Interacting Multiple Model Method for Small Bird Tracking on Lidar System
Lidar presents a promising solution for bird surveillance in airport environments. However, the low observation refresh rate of Lidar poses challenges for tracking bird targets. To address this problem, we propose a gated recurrent unit (GRU)-based interacting multiple model (IMM) approach for tracking bird targets at low sampling frequencies. The proposed method constructs various GRU-based motion models to extract different motion patterns and to give different predictions of target trajectory in place of traditional target moving models and uses an interacting multiple model mechanism to dynamically select the most suitable GRU-based motion model for trajectory prediction and tracking. In order to fuse the GRU-based motion model and IMM, the approximation state transfer matrix method is proposed to transform the prediction of GRU-based network into an explicit state transfer model, which enables the calculation of the models’ probability. The simulation carried out on an open bird trajectory dataset proves that our method outperforms classical tracking methods at low refresh rates with at least 26% improvement in tracking error. The results show that the proposed method is effective for tracking small bird targets based on Lidar systems, as well as for other low-refresh-rate tracking systems.