In safety inspection tasks in building construction sites, 3D point cloud object detection is a key technology for safety inspection robots to perceive construction workers, construction vehicles, and surrounding obstacles. However, sparse low-channel Light Detection and Ranging (LiDAR) point clouds, dynamic-object interference, and deployment on resource-constrained robotic platforms make it difficult to improve detection accuracy without compromising detector-level inference efficiency. To address this problem, this study proposes CRobot-SECOND, a 3D point cloud object detection method that combines point cloud input enhancement with lightweight backbone optimization. First, a point cloud fusion method based on ground constraints and dynamic object compensation is constructed to improve point cloud density and geometric completeness while suppressing ghosting artifacts. Then, the Partial Convolution (PConv) module and the Simple Parameter-Free Attention Module (SimAM) are introduced into the two-dimensional Bird’s-Eye-View (BEV) backbone of Sparsely Embedded Convolutional Detection (SECOND) to reduce redundant computation and enhance key-region feature representation. Finally, a construction-site point cloud dataset containing construction workers and construction vehicles is established, and comparative, ablation, and multi-seed experiments are conducted. The experimental results show that the proposed fusion method improves point cloud density while maintaining high fusion quality. Across five independent training seeds, CRobot-SECOND achieves 3D average precision (AP) values of (57.77 ± 0.50)% for construction workers and (82.84 ± 0.37)% for construction vehicles, compared with (53.09 ± 0.44)% and (81.08 ± 0.32)% for SECOND, respectively, where the results are reported as mean ± standard deviation. The category-averaged 3D AP increases from (67.08 ± 0.35)% to (70.31 ± 0.43)%, with a mean paired improvement of 3.22 ± 0.10 percentage points. The detector-only network inference speed increases from 75.19 Hz to 92.08 Hz. Overall, these results show that coordinated point cloud input enhancement and lightweight detector optimization improve frame-level 3D detection of construction workers and construction vehicles under the collected construction-site conditions without increasing model scale. These frame-level detection results provide a basis for exploring CRobot-SECOND as a perception component for construction-site safety monitoring, robot path planning, and obstacle avoidance.
| [1] |
Liu Y, Alias AH, Haron NA, Bakar NA, Wang H. Robotics in the construction sector: trends, advances, and challenges. Journal of Intelligent & Robotic Systems, 2024, 1102Article 72.
|
| [2] |
Farrokh Ghatte H, Nazari A. Prediction of rice husk ash-based SCC compressive strength: Data-driven framework. Civil Engineering and Applied Solutions, 2025, 131728.
|
| [3] |
Sabetifar H, Mousavimehr M, Aminian A, Nematzadeh M. Residual axial performance of PET/rubber-modified concrete confined with CFRP strips after thermal exposure: Experimental and theoretical analysis. Developments in the Built Environment, 2026, 25: 100834.
|
| [4] |
Sasania F, Nematzadeh M, Asadi M, Aminian A. Thermo-mechanical behavior of high-strength concrete with nylon granule aggregates: Experimental evaluation and predictive analysis. Civil Engineering and Applied Solutions, 2025, 11135.
|
| [5] |
Nemati M, Aminian A, Rahimi S, Nematzadeh M, Jafarzadeh-Taleshi M, Thai H-T. Compressive behavior of prestressed SFRCFST stub columns after heating: Effect of fresh concrete compression technique. Case Studies in Construction Materials, 2025, 23: e04968.
|
| [6] |
Valizadeh Kiamahalleh M, Khorshidi-Mianaei A, Safari A. Improvement of recycled concrete aggregate properties by polyvinyl alcohol. Civil Engineering and Applied Solutions, 2026, 224959.
|
| [7] |
Norouzi A, Naghipour M. Evaluation of the static behavior of WPC-GFRP sandwich panels: An experimental, theoretical, and numerical study. Civil Engineering and Applied Solutions, 2026, 22127.
|
| [8] |
Liu Y, Bin Alias AH, Haron NA, Abu Bakar N, Wang H. Exploring three pillars of construction robotics via dual-track quantitative analysis. Automation in Construction, 2024, 162: Article 105391.
|
| [9] |
Hu D, Gan VJL, Yin C. Robot-assisted mobile scanning for automated 3D reconstruction and point cloud semantic segmentation of building interiors. Automation in Construction, 2023, 152: 104949.
|
| [10] |
Halder S, Afsari K, Chiou E, Patrick R, Hamed KA. Construction inspection & monitoring with quadruped robots in future human-robot teaming: A preliminary study. Journal of Building Engineering, 2023, 65: 105814.
|
| [11] |
Chen Z, Song C, Wang B, Tao X, Zhang X, Lin F, Cheng JCP. Automated reality capture for indoor inspection using BIM and a multi-sensor quadruped robot. Automation in Construction, 2025, 170: 105930.
|
| [12] |
Feng C, Yu J, Li J, Wang K, Wang B, Wu Y. Evaluation of LiDAR SLAM algorithms for construction robots in large public construction sites. Low-carbon Materials and Green Construction, 2025, 3128.
|
| [13] |
Tuomisto J, Zheng Y, Chauhan I, Seppanen O, Pajarinen J. Automating on-site object inspection with a quadruped robot and BIM. Automation in Construction, 2026, 181: Article 106571.
|
| [14] |
Kim J, Kim J, Koo N, Kim H. Automating scaffold safety inspections using semantic analysis of 3D point clouds. Automation in Construction, 2024, 166: 105603.
|
| [15] |
Yue H, Wang Q, Zhao H, Zeng N, Tan Y. Deep learning applications for point clouds in the construction industry. Automation in Construction, 2024, 168: 105769.
|
| [16] |
Mao J, Shi S, Wang X, Li H. 3D object detection for autonomous driving: A comprehensive survey. International Journal of Computer Vision, 2023, 131819091963.
|
| [17] |
Lu B, Sun Y, Yang Z, Song R, Jiang H, Liu Y. HRNet: 3D object detection network for point cloud with hierarchical refinement. Pattern Recognition, 2024, 149: Article 110254.
|
| [18] |
Tang Q, Yang M, Wang Z, Dong W, Liu Y. Boundary points guided 3D object detection for point clouds. Applied Soft Computing, 2024, 165: Article 112117.
|
| [19] |
Rauch L, Braml T. Rohbau3D: a shell construction site 3D point cloud dataset. Scientific Data, 2025, 121Article 1478.
|
| [20] |
Park G, Koh J, Kim J, Moon J, Choi JW. LiDAR-based 3D temporal object detection via motion-aware LiDAR feature fusion. Sensors, 2024, 2414Article 4667.
|
| [21] |
Wang Y, Lyu R, Ouyang J, Wang Z, Xie X, Chen H. DynaLOAM: Robust LiDAR odometry and mapping in dynamic environments. Autonomous Robots, 2025, 494Article 29.
|
| [22] |
Liu H, Du J, Zhang Y, Zhang H, Zeng J. PVConvNet: Pixel-voxel sparse convolution for multimodal 3D object detection. Pattern Recognition, 2024, 149: Article 110284.
|
| [23] |
Li Y, Li Q, Gao C, Gao S, Wu H, Liu R. PFENet: Towards precise feature extraction from sparse point cloud for 3D object detection. Neural Networks, 2025, 185: Article 107144.
|
| [24] |
Chen W, Jiang X, Ji Z, Liu H, Zhu G, Armendariz-Inigo JE, Yin S. Rfpillarnext: Geometry-aware 3D object detection for dynamic LiDAR point clouds. The Visual Computer, 2025, 41141202512036.
|
| [25] |
Meng X, Zhou Y, Zhang M, Wang C, Lv J, Kim J, Wang S. MFTFF: Multi-frame 3D object detection with temporal feature fusion. Journal of Supercomputing, 2025, 818Article 963.
|
| [26] |
Xu N, Qin R, Song S. Point cloud registration for LiDAR and photogrammetric data: A critical synthesis and performance analysis on classic and deep learning algorithms. ISPRS Open Journal of Photogrammetry and Remote Sensing, 2023, 8: 100032-100032.
|
| [27] |
Wang Y, Zhou P, Geng G, An L, Zhang Q. End-to-end point cloud registration with transformer. Artificial Intelligence Review, 2024, 58126.
|
| [28] |
Peng H, Zhao Z, Wang L. A review of dynamic object filtering in SLAM based on 3D LiDAR. Sensors, 2024, 242Article 645.
|
| [29] |
Yue Y, Qi H, Deng Y, Li J, Liang H, Miao J. Infrastructure-side point cloud object detection via multi-frame aggregation and multi-scale fusion. IEEE Transactions on Intelligent Transportation Systems, 2025, 261703713.
|
| [30] |
Qian R, Lai X, Li X. 3D object detection for autonomous driving: A survey. Pattern Recognition, 2022, 130: 108796.
|
| [31] |
Liu Y, Yang Z, Tong J, Yang J, Peng J, Zhang L, Cheng W. ET-PointPillars: Improved PointPillars for 3D object detection based on optimized voxel downsampling. Machine Vision and Applications, 2024, 35356.
|
| [32] |
Sun L, Li Y, Qin W. PEPillar: A point-enhanced pillar network for efficient 3D object detection in autonomous driving. The Visual Computer, 2025, 41317771788.
|
| [33] |
Wang J, Yu Y, Cheng J, Li J, Tang J. PillarBAPI: Enhancing pillar-based 3D object detection through attentive pseudo-image feature extraction. Multimedia Systems, 2025, 313Article 263.
|
| [34] |
Wu P, Gu L, Yan X, Xie H, Wang FL, Cheng G, Wei M. PV-RCNN++: Semantical point-voxel feature interaction for 3D object detection. The Visual Computer, 2023, 39624252440.
|
| [35] |
Fan L, Cao J, Liu X, Li X, Deng L, Sun H, Peng Y. Voxel self-attention and center-point for 3D object detector. iScience, 2024, 279110759.
|
| [36] |
Zhang W, Chen G, Wang H, Yang L, Sun T. EFMF-pillars: 3D object detection based on enhanced features and multi-scale fusion. Eurasip Journal on Advances in Signal Processing, 2024, 20241Article 90.
|
| [37] |
Xue C, Xia Y, Wu M, Chen Z, Cheng F, Yun L. EL-YOLO: An efficient and lightweight low-altitude aerial objects detector for onboard applications. Expert Systems with Applications, 2024, 256: 124848.
|
| [38] |
Yang S, Chen L, Wang J, Jin W, Yu Y. A novel lightweight object detection network with attention modules and hierarchical feature pyramid. Symmetry, 2023, 15112080.
|
| [39] |
Guo Y, He Y, Zhang H, Ma J. YOLO-CAM: A lightweight UAV object detector with combined attention mechanism for small targets. Remote Sensing, 2025, 17213575.
|
| [40] |
Zou Z, Lang H, Lou Y, Lu J. Plane-based global registration for pavement 3D reconstruction using hybrid solid-state LiDAR point cloud. Automation in Construction, 2023, 152: Article 104907.
|
| [41] |
Peng Y, Lin S, Wu H, Cao G. Point cloud registration based on fast point feature histogram descriptors for 3D reconstruction of trees. Remote Sensing, 2023, 1515Article 3775.
|
| [42] |
Lu G, Zhang W, Wang Z. Optimizing depthwise separable convolution operations on GPUs. IEEE Transactions on Parallel and Distributed Systems, 2022, 3317087.
|
| [43] |
Ping P, Yang Z, Tao L, Shi Q, Ding W. BEV-TinySpotter: A novel BEV perception method considering multi-dimensional feature fusion of small target. Information Fusion, 2025, 116: Article 102793.
|
| [44] |
Sun T, Hao Y, Huang S, Savarese S, Schindler K, Pollefeys M, Armeni I. Nothing stands still: A spatiotemporal benchmark on 3D point cloud registration under large geometric and temporal change. ISPRS Journal of Photogrammetry and Remote Sensing, 2025, 220: 799-823.
|
| [45] |
Qian R, Lai X, Li X. BADet: Boundary-aware 3D object detection from point clouds. Pattern Recognition, 2022, 125: Article 108524.
|
Funding
Key Research and Development Program of Shaanxi(2025SF-GJHX2-14)
Science and Technology Project of Shaanxi Provincial Department of Housing and Urban-Rural Development(2025-K05)
Rights & permissions
The Author(s)
PDF
