基于三维声纳点云的水下沉井地形探测技术

Underwater caisson topography detection technology based on 3D sonar point clouds

为了解决水下沉井地形监测难题,给出了一种基于三维扫描声纳的地形监测方法以及地形点云的高精度快速处理方法。首先,结合沉井特点,给出了一种悬挂式三维扫描声纳扫描沉井水下地形方法;然后,结合沉井形状及点云特点,给出由降采样点云抽稀法、基于欧式距离的区域自生长算法的点云去噪方法和基于K-means++聚类算法的沉井水下地形分割方法等组成的水下地形点云处理方法,实现了基于声纳点云的地形点云快速获取及水下地形恢复。实验结果表明,该方法可有效用于沉井水下地形探测。

In order to solve the problem of underwater caisson terrain monitoring, a terrain monitoring method based on 3 D scanning sonar and a high precision and fast processing method of terrain point cloud are presented in this paper. Firstly, a scanning method of hanging 3 D scanning sonar for underwater topography of caisson is put forward, combining with the characteristics of caisson. And then, based on the shape of the caisson and the characteristics of point cloud, the processing method of underwater terrain point cloud is presented, which consists of down sampling point cloud diluted method, the point cloud denoising method based on the area self-growth algorithm of Euclidean distance, and the underwater terrain segmentation method of caisson based on K-means++ clustering algorithm. By doing this, the quick acquisition of terrain point cloud and the recovery of underwater terrain based on the sonar point cloud are implemented. The experimental results show that the proposed method can be used effectively for underwater topography detection of caisson.