张目宁,杨鲲,吴永亭,柳义成,曹玉芬,阳凡林.用于海底目标识别与底质分类的多波束水体波形预处理[J].海洋通报,2021,(1):
用于海底目标识别与底质分类的多波束水体波形预处理
Multi-beam water columndata processing method for target recognition and classification of seabed sediments
投稿时间:2020-07-14  修订日期:2020-11-25
DOI:
中文关键词:  多波束水体数据  波形去噪  分区波形拟合  双指数拟合
英文关键词:multi-beam water columndata  waveform fitting  zonal waveform fitting  double exponential fitting
基金项目:国家自然基金重点项目 (41930535);国家重点研发计划 (2018YFF0212203;2017YFC1405006;2018YFC1405900);山东科技大学科研创新团队支持计划 (2019TDJH103)
作者单位E-mail
张目宁 山东科技大学 测绘科学与工程学院山东 青岛 266590 自然资源部海洋测绘重点实验室山东 青岛 266590 15726204801@163.com 
杨鲲 交通运输部天津水运工程科学研究院天津 300000  
吴永亭 自然资源部第一海洋研究所山东 青岛 266061  
柳义成 交通运输部天津水运工程科学研究院天津 300000  
曹玉芬 交通运输部天津水运工程科学研究院天津 300000  
阳凡林 山东科技大学 测绘科学与工程学院山东 青岛 266590 自然资源部海洋测绘重点实验室山东 青岛 266590 yang723@163.com 
摘要点击次数: 172
全文下载次数: 90
中文摘要:
      多波束水体数据是多波束系统获取的最原始数据,记录了波束从发射到接收整个过程全部的反向散射强度信息,可以为目标识别、水下栖息环境探测等提供重要的数据支撑。目前,针对多波束水体强度时间序列所表现的波形信息的处理及 研究仍处于起步阶段,另外水体波形数据易受噪声影响,且存在明显的入射角效应问题,对此,本文提出了一种基于分区异构的多波束水体波形拟合算法。首先,根据不同波束入射角范围的水体波形特性,将水体数据划分为 3 个区域;然后利用不 同函数 (中央波束区域—双指数函数、漫反射区域—广义高斯与线性函数叠加、边缘波束区域—高斯与多项式叠加) 分别对不同分区的反向散射强度波形进行拟合。采用台湾海峡的多波束水体数据进行验证,结果表明:不同分区拟合相关系数及拟 合优度均达到 0.95 以上,相比简单函数拟合,均方根误差由 3.39 dB 降到 1.5 dB 以下,达到了较好的拟合效果,可为多波束水体目标识别和海底分类提供参考。
英文摘要:
      Multi-beam water column data is the most original data obtained by multi-beam echo soundersystem. It records all the backscattering intensity information of the entire process from transmission to reception, providing important support for target recognition and underwater habitat detection. At present, the processing and research of the waveform information represented bymulti-beam water columntimeseriesintensity is still in its infancy. In addition, the waveform of water column data are easily affected by noise and have obvious incident angle effect. This paper proposes a multi-beam water column data waveform fitting algorithm based on partitioned heterogeneity. First of all, according to the water column waveform characteristics of different beam incidence angle ranges, the water column data is divided into 3 areas. Secondly, the backscattering intensities of different partitions are respectively fitted with different functions (central beam area-double exponential function,diffuse reflection area-superposition of generalized Gaussian function and linear function, edge beam area -Gaussian and polynomial superposition). The proposed algorithm was verified by using multi -beam water data collected in the Taiwan Strait. Experiments show that the correlation coefficient and goodness-of-fit are above 0.95. Compared with simple function fitting, the root-mean-square error is reduced from 3.39 dB to less than 1.5 dB, suggesting a good fit. The algorithm provides an effective basis for the extraction of waveform features of the water column, and has certain application potential in the identification of bottom targets and the classification of seabed.
查看全文    下载PDF阅读器
关闭