控制与信息技术Issue(6)：42-47,6.DOI:10.13889/j.issn.2096-5427.2025.06.005
激光点云与BIM点云联合下的大空间场景异常SSA-BP识别技术
SSA-BP Anomaly Detection Approach for Large-Space Scenes Using Combined Laser and BIM Point Clouds
钱国成 1宋金闻 1陈宇 1史昊东1
作者信息
- 1. 中国水利水电第十四工程局有限公司,云南 昆明 650200
- 折叠
摘要
Abstract
To obtain more comprehensive point cloud information for large space scenes and achieve accurate detection of anomalies particularly in large-space building scenes,this paper proposes a Sparrow Search Algorithm-Back Propagation(SSA-BP)anomaly detection approach that combines laser and Building Information Modeling(BIM)point clouds.Initial registration of the laser and BIM point clouds in large-space scenes utilizes the Principal Component Analysis-Iterative Closest Point(PCA-ICP)method that accounts for principal directions.The Sparrow Search Algorithm(SSA)is then used to optimize the parameters of a BP neural network.The resulting SSA-BP automatically extracts anomaly features from point clouds in large-space scenes to perform anomaly detection.Experimental results demonstrate the approach's effectiveness for high-precision joint registration of laser and BIM point clouds and for accurate anomaly detection in large-space scenes.关键词
激光点云/BIM点云/大空间场景/异常识别/麻雀搜索算法/BP神经网络Key words
laser point cloud/BIM point cloud/large-space scene/anomaly detection/sparrow search algorithm/BP neural network分类
信息技术与安全科学引用本文复制引用
钱国成,宋金闻,陈宇,史昊东..激光点云与BIM点云联合下的大空间场景异常SSA-BP识别技术[J].控制与信息技术,2025,(6):42-47,6.
