基于极点特征聚类的智能建筑混凝土墙体裂缝识别方法  

Method for Identifying Cracks in Concrete Wall of Intelligent Building Based on Pole Feature Clustering

作　　者：赵青羽 ZHAO Qing-yu(The First Construction Engineering Company Ltd.of China Construction Second Engineering Bureau)

机构地区：[1]中建二局第一建筑工程有限公司

出　　处：《智能建筑与智慧城市》2022年第11期72-74,共3页Intelligent Building & Smart City

摘　　要：为精准识别墙体裂缝,文章研究基于极点特征聚类的智能建筑混凝土墙体裂缝识别方法。利用改进近邻保留嵌入算法,降维处理混凝土墙体回波信号;求解信号的冲激响应,获取回波信号的极点特征;基于Grassmann流行距离度量的谱聚类算法,聚类回波信号数据点,输出墙体裂缝识别结果。实验证明:该方法可精准识别墙体裂缝。In order to accurately identify wall cracks,the method of identifying cracks in concrete wall of intelligent building based on pole feature clustering is studied.The improved nearest neighbor preserving embedding algorithm is used to reduce the dimension of concrete wall echo signal.The impulse response of the signal is solved and the pole characteristics of the echo signal are obtained.Spectral clustering algorithm based on Grassmann popular distance measure is used to cluster the echo signal data points and output wall crack identification results.Experiments show that this method can accurately identify wall cracks.

关 键 词：极点特征 智能建筑 混凝土墙体 裂缝识别 Grassmann流行距离 谱聚类算法 

分 类 号：TN957.52[电子电信—信号与信息处理] TU17[电子电信—信息与通信工程]