基于最近点迭代的水下三维声图增强算法  

An Image Enhancement of Underwater Three-dimensional Sonar Image Based on ICP

作　　者：曾腾 王朋[1,2] 黄海宁[1,2] 王冠群 张武[1,2] ZENG Teng;WANG Peng;HUANG Haining;WANG Guanqun;ZHANG Wu(Key Laboratory of Science and Technology on Advanced Underwater Acoustic Signal Processing,Chinese Academy of Science,Beijing,100190,China;Laboratory of Underwater Acoustic Signal Processing,Institute of Acoustics,Chinese Academy of Sciences,Beijing,100190,China)

机构地区：[1]中国科学院先进水下信息技术重点实验室,北京100190 [2]中国科学院声学研究所水下信息技术实验室,北京100190

出　　处：《网络新媒体技术》2024年第3期19-25,35,共8页Network New Media Technology

摘　　要：由于复杂的水下环境和声呐硬件的限制,三维成像声呐(3DIS)图像质量较差,表现为图像目标表面点的稀疏和缺失。为了提高图像质量,我们提出了一种基于配准和融合的目标增强方法。迭代最邻近点(ICP)算法是经典的点云配准算法,是一种用于配准2个或多个点云数据集的迭代优化方法。ICP方法可直接用于点云数据,具有实施简单、迭代速度快的优点。该算法通过对齐和融合的方法叠加多帧点云来增强点云,可以改善目标点云稀疏和缺失的问题。实验结果表明,我们提出的增强方法能有效增强声呐图像,提高三维声呐图像的质量和可理解性。The 3-D imaging sonar(3DIS)image quality is poor,manifested as sparse and missing of target surface points,owing to complex underwater environment and hardware limitation of sonar.To improve the quality of the image,we propose a target enhancement method based on registration and merge.The ICP(Iterative Closet Point)algorithm is a classical point cloud alignment algorithm,whose is an iterative optimization method for aligning two or more point cloud datasets.The ICP method can be used directly with point cloud data and has the advantage of being simple to implement and fast to iterate.The algorithm in the paper enhances the point cloud by superimposing multi-frame point clouds by means of alignment and fusion,which can improve the problem of sparse and missing target point clouds.Experimental results demonstrate that our proposed enhancement method can effectively enhance the object result,improve the quality and understandability of the 3DIS image.

关 键 词：图像增强 三维成像声呐 图像拼接 点云 最近点迭代 

分 类 号：U666.7[交通运输工程—船舶及航道工程] TP391.41[交通运输工程—船舶与海洋工程]