基于轮廓线聚类分析的船舶超吃水检测  

Draught Monitoring Based on Contour Cluster Analysis

作　　者：谢磊[1] 薛双飞 黄立[1] 包竹 XIE Lei;XUE Shuangfei;HUANG Li;BAO Zhu(National Engineering Research Center for Water Transport Safety, Wuhan University of Technology,Wuhan 430063,China)

机构地区：[1]武汉理工大学国家水运安全工程技术研究中心,武汉430063

出　　处：《中国航海》2018年第1期60-63,108,共5页Navigation of China

基　　金：国家自然科学基金(51479158);交通运输部应用基础研究项目(2013329811220)

摘　　要：针对在船舶吃水检测中采用人工登船测量法工作量大、实施效率低等问题,归纳分析现有的船舶吃水状态自动检测技术。从检测的准确性和可行性出发,采用岸基检测法获取船舶激光点云数据;通过一种基于聚类分析的船舶水面轮廓重构方法剔除激光检测产生的大量噪声点,完成船舶水上部分三维重构;结合船舶静态信息推算出船舶吃水状态。试验结果证明,该方法能快速准确地获取船舶三维重构图,可作为船舶超吃水检测手段。Manually checking the draught of a ship can be inconvenient and inefficient.The automatic monitoring technology based on contour analysis is proposed which produces 3D image of the hull with the point cloud from the laser scanner.With the image the system can find the draught of the ship through checking the contour of the above water part of the hill against its drawing.The cluster algorithm is used for getting rid of measurement noises.Tests prove that the design works satisfactory with high accuracy and quick response.

关 键 词：激光点云 吃水检测 模糊C-均值聚类 三维重构 

分 类 号：U665.26[交通运输工程—船舶及航道工程]