【作者】 张嘉维； 刘爱荣； 陈炳聪；

【Author】 ZHANG Jia-wei;LIU Ai-rong;CHEN Bing-cong;Research Center of Wind Engineering and Engineering Vibration;

【机构】 广州大学风工程与工程振动研究中心；

【摘要】 针对水下混凝土桩检测中环境复杂、传统光学成像受和检测效果不直观等问题,本文提出基于多波束图像声纳的水下缺陷桩三维重建方法。通过水下机器人搭载二维图像声呐,在环绕桩基的多个方位与不同入水深度采集声呐图像,利用特征点比对与轮廓平滑连接构建三维模型。试验结果表明:该方法对凹陷、孔洞等缺陷的重建具有较高可靠性。剥落缺陷因声呐精度限制存在一定误差,但在实际工程较大深度下仍具可行性。以上研究表明本文所提方法可有效对水下缺陷桩进行三维重建。更多还原

【Abstract】 Aiming at the issues of complex environment, limitations of traditional optical imaging, and nonintuitive inspection effects in underwater concrete pile detection, this paper proposes a three-dimensional reconstruction method for underwater defective piles based on multibeam imaging sonar. By using an underwater robot equipped with two-dimensional imaging sonar, sonar images are acquired from multiple azimuths around the pile foundation and at different water depths. A 3D model is constructed through feature point matching and contour smooth connection. Experimental results show that this method has high reliability in reconstructing defects such as depressions and holes. Although spalling defects have certain errors due to sonar precision limitations, the method remains feasible at larger depths in practical engineering. The above research indicates that the proposed method can effectively perform 3D reconstruction of underwater defective piles.更多还原