宽带多波束渔用声呐显控系统设计及关键技术研究  

Design and key technology research on display and control system of wideband multibeam fishery sonar

作　　者：宗艳梅 唐学大 楚树坡 谌志新 李国栋[1,2] ZONG Yanmei;TANG Xueda;CHU Shupo;CHEN Zhixin;LI Guodong(Qingdao Marine Science and Technology Center,Qingdao Shandong266237,China;Fishery Machinery and Instrument Research Institute,Chinese Academy of Fishery Sciences,Shanghai200092,China)

机构地区：[1]青岛海洋科技中心,山东青岛266237 [2]中国水产科学研究院渔业机械仪器研究所,上海200092

出　　处：《海洋渔业》2023年第4期500-512,共13页Marine Fisheries

基　　金：国家重点研发计划(2019YFD0901403);国家自然科学基金项目(32073026);三亚崖州湾科技城科研项目(SKJC-2020-01-013);中国水产科学研究院中央级公益性科研院所基本科研业务费专项基金(2020TD83)。

摘　　要：针对宽带多波束渔用声呐实时探测数据量大、处理数据慢等特点,设计了一种宽带多波束渔用声呐显控系统。该系统采用控制层、显示层和数据层3层架构设计,运用多线程技术实时处理数据,并采用多种插值方法实现声呐图像可视化。经过试验对比分析,声呐图像采用双立方插值算法其PSNR值为29.36 dB,可视化效果较好。为进一步提高声呐图像分辨率,采用中值滤波、均值滤波、双边滤波及高斯滤波算法对声呐图像进行去噪效果对比,其PSNR值分别为29.67 dB、28.66 dB、27.64 dB、28.09 dB,结果表明,中值滤波算法去噪效果较好。通过反复试验和论证,本系统可实时高效探测目标信息,可作为宽带多波束渔用声呐设备的显控软件,具有较强的实用价值。Wideband multi beam fishery sonar can be used for long-distance and high-resolution detection.With higher efficiency in detecting fish school,it has an important application in accurate detection and efficient fishing of marine fisheries.Aiming at the characteristics of large amount of real-time detection data and slow data processing of wideband multibeam fishery sonar,the technology research of real-time processing of sonar data and improving the resolution of sonar image was carried out,and a display and control system of wideband multibeam fishery sonar was designed.The system adopted three-layer architecture design of control layer,display layer and data layer,and used multithreading technology to process data,store data and display sonar images in real time.The system adopted modular design idea and was divided into four modules according to functions:task bar module,control parameter configuration module,display parameter configuration module and sonar image display module.This study verified and compared by simulation test and anechoic pool test.The display and control system used three algorithms:the nearest neighbor interpolation,bilinear interpolation and bicubic interpolation to realize sonar image visualization.After comparative analysis of simulation experiment and pool experiment,the advantages and disadvantages of three interpolation algorithms were quantitatively analyzed by using peak signal-to-noise ratio.The peak signal-to-noise ratio of sonar image in pool experiment was 26.01 dB,28.07 dB and 29.36 dB,respectively.The results showed that compared with the nearest neighbor interpolation and bilinear interpolation,the peak signal-to-noise ratio of sonar image processed by bicubic interpolation was the highest and the visualization effect was better.In order to further improve the resolution of sonar image and reduce the noise in sonar image,median filtering,mean filtering,bilateral filtering and Gaussian filtering were used to denoise sonar image.Through comparative analysis of simulation test

关 键 词：宽带多波束技术 声呐图像插值 声呐图像滤波 渔用声呐显控系统 

分 类 号：S972.61[农业科学—捕捞与储运]