RepDNet:A re-parameterization despeckling network for autonomous underwater side-scan sonar imaging with prior-knowledge customized convolution

Side-scan sonar(SSS)is now a prevalent instrument for large-scale seafloor topography measurements,deployable on an autonomous underwater vehicle(AUV)to execute fully automated underwater acoustic scanning imaging along a predetermined trajectory.However,SSS images often suffer from speckle noise caused by mutual interference between echoes,and limited AUV computational resources further hinder noise suppression.Existing approaches for SSS image processing and speckle noise reduction rely heavily on complex network structures and fail to combine the benefits of deep learning and domain knowledge.To address the problem,Rep DNet,a novel and effective despeckling convolutional neural network is proposed.Rep DNet introduces two re-parameterized blocks:the Pixel Smoothing Block(PSB)and Edge Enhancement Block(EEB),preserving edge information while attenuating speckle noise.During training,PSB and EEB manifest as double-layered multi-branch structures,integrating first-order and secondorder derivatives and smoothing functions.During inference,the branches are re-parameterized into a 3×3 convolution,enabling efficient inference without sacrificing accuracy.Rep DNet comprises three computational operations:3×3 convolution,element-wise summation and Rectified Linear Unit activation.Evaluations on benchmark datasets,a real SSS dataset and Data collected at Lake Mulan aestablish Rep DNet as a well-balanced network,meeting the AUV computational constraints in terms of performance and latency.

Application Research of Sonar Detection Method in Melting Exploration at the Bottom of Piles

Karst landforms are widely distributed in China,and are most common in Yunnan,Guizhou and Guangxi.If the development of karst caves at the bottom of the piles cannot be accurately ascertained before the construction of bridge pile foundations,accidents such as hole collapse,slurry leakage,and drill sticking will easily occur.In this paper,the principle and method of sonar detection for detecting karst caves at the bottom of bridge piles was introduced,and the sonar detection data and the cave situation at the bottom of the pile during the construction process in combination with the case of Yunnan Zhenguo Highway Project was analyzed,which verifies the practicability and reliability of sonar detection method reliability.

DcNet: Dilated Convolutional Neural Networks for Side-Scan Sonar Image Semantic Segmentation

In ocean explorations,side-scan sonar(SSS)plays a very important role and can quickly depict seabed topography.As-sembling the SSS to an autonomous underwater vehicle(AUV)and performing semantic segmentation of an SSS image in real time can realize online submarine geomorphology or target recognition,which is conducive to submarine detection.However,because of the complexity of the marine environment,various noises in the ocean pollute the sonar image,which also encounters the intensity inhomogeneity problem.In this paper,we propose a novel neural network architecture named dilated convolutional neural network(DcNet)that can run in real time while addressing the above-mentioned issues and providing accurate semantic segmentation.The proposed architecture presents an encoder-decoder network to gradually reduce the spatial dimension of the input image and recover the details of the target,respectively.The core of our network is a novel block connection named DCblock,which mainly uses dilated convolution and depthwise separable convolution between the encoder and decoder to attain more context while still retaining high accuracy.Furthermore,our proposed method performs a super-resolution reconstruction to enlarge the dataset with high-quality im-ages.We compared our network to other common semantic segmentation networks performed on an NVIDIA Jetson TX2 using our sonar image datasets.Experimental results show that while the inference speed of the proposed network significantly outperforms state-of-the-art architectures,the accuracy of our method is still comparable,which indicates its potential applications not only in AUVs equipped with SSS but also in marine exploration.

YOLOv5-Based Seabed Sediment Recognition Method for Side-Scan Sonar Imagery

Seabed sediment recognition is vital for the exploitation of marine resources.Side-scan sonar(SSS)is an excellent tool for acquiring the imagery of seafloor topography.Combined with ocean surface sampling,it provides detailed and accurate images of marine substrate features.Most of the processing of SSS imagery works around limited sampling stations and requires manual interpretation to complete the classification of seabed sediment imagery.In complex sea areas,with manual interpretation,small targets are often lost due to a large amount of information.To date,studies related to the automatic recognition of seabed sediments are still few.This paper proposes a seabed sediment recognition method based on You Only Look Once version 5 and SSS imagery to perform real-time sedi-ment classification and localization for accuracy,particularly on small targets and faster speeds.We used methods such as changing the dataset size,epoch,and optimizer and adding multiscale training to overcome the challenges of having a small sample and a low accuracy.With these methods,we improved the results on mean average precision by 8.98%and F1 score by 11.12%compared with the original method.In addition,the detection speed was approximately 100 frames per second,which is faster than that of previous methods.This speed enabled us to achieve real-time seabed sediment recognition from SSS imagery.

Multi-beam Sonar and Side-scan Sonar Image Co-registering and Fusing

Multi-beam Sonar and Side-scan Sonar compensate each other. In order to fully utilize all information, it is necessary to fuse two kinds of image and data. And the image co-registration is an important and complicated job before fusion. This paper suggests combining bathymetric data with intensity image, obtaining the characteristic points through the minimal angles of lines, and then deciding the corresponding image points by the maximal correlate coefficient in searching space. Finally, the second order polynomial is applied to the deformation model. After the images have been co-registered, Wavelet is used to fuse the images. It is shown that this algorithm can be used in the flat seafloor or the isotropic seabed. Verification is made in the paper with the observed data.

A Research on Road and Bridge Detection Based on Test Detection Technology

The existence of karst cave at the bottom of bored piles has a great impact on projects under construction and the surrounding buildings.Since bored piles require slurry wall protection,the current geophysical exploration method cannot effectively detect the karst cave at the bottom of the piles in the slurry.Combined with the characteristics of stress wave propagation,the sonar detection method is proposed.JL sonar detector can realize the transmission and acquisition of on-site sonar signals.This method makes full use of the mud conditions of bored cast-in-place piles,and the development of karst caves can be tracked and detected within 10 meters at the pile bottom during the drilling process.It has several advantages,including low cost,high speed,and high precision.This paper verifies the application of sonar detection technology in practical engineering through specific engineering cases.The research results put forward a new solution for cave exploration in karst areas,especially in liquid environment.

Sedimentary processes in Zenisu deep-sea channel revealed by side-scan imagery

Side-scan sonar data collected by Cruises 99-09 Leg 2 and 00-06 Leg l of R/V Yokosuka were used to reveal the sedimentary processes in Zenisu deep-sea channel. The middle and lower segments of the channel are rich in turbidite and other debrite deposits. By high-resolution imaging, three sedimentary processes were distinguished with distinct acoustic features. 1. Slumps and slides occur with contrasting backscatter, rough surface textures, blockings, and acoustic shadows at headwalls. They are very extensive and often in lobate form downslope. 2. Debris flow has uniform, general medium backscatter, sometimes showing marbling/lineation in lobate form. 3. Turbidity current is characterized by low backscatter confined to the channel as acoustic signal is attenuated. Regional tectonics must be the dominating factor that controls deposition pattern in this area.

基于主动识别声呐的养殖塘南美白对虾探测与初步分析

为实现高效、快速和准确的虾类识别和行为观测,提出一种基于主动识别声呐(DIDSON)的养殖塘南美白对虾探测和行为分析方法。该方法利用无人艇搭载主动识别声呐(DIDSON)对养殖塘设置的8个观测点进行连续定点观测,利用基于ECHOVIEW构建的虾类识别计数模型对采集的图像数据进行识别提取,通过目标识别、目标提取、通量分析以及方向分析等方法获得养殖塘南美白对虾的初步行为特征。试验在8个观测点观测的虾群数量极值分别为251只和208只,均值为234只,单位时间内的虾群通量区间为[108,131]只/(min·m^(2)),不同点位虾群正向游塘的占比均大于85%,通量变化较小。结果显示,该方法可以观测到养殖塘内虾群存在规律性的游塘行为,相较于传统的水下视觉观察和被动声呐调查,有效解决了养殖塘中虾群行为观测的难题,为制定更加高效的虾塘饲养和管理方案提供科学的数据支持。

基于轻量化YOLOv7算法的侧扫声纳图像沉船检测

针对现有的侧扫声纳图像水下沉船检测方法存在检测速度慢,传统的YOLOv5算法存在的漏检的问题,提出基于轻量化YOLOv7算法的水下沉船检测改进方法。首先,通过随机翻转、随机噪声等操作扩充沉船图像的样本数量;然后,引入迁移学习策略,将在COCO数据集上学习到的权重迁移到沉船检测的YOLOv7网络中;其次,改进模型损失函数中惩罚项的计算方式,提升收敛速度;最后在YOLOv7网络中引入FasterNet结构,减少模型的参数量和计算复杂度,降低模型对硬件的需求,达到轻量化模型的目的。实验结果表明,改进方法较原始YOLOv7算法在类平均精度值(mAP值)上提升了4.75%,检测速度也由原来的0.0218秒/帧提升到0.0179秒/帧,证明了改进方法的工程应用价值。

基于声纳搜索累积探测概率的平台路径优化方法

针对面向移动目标的移动搜索平台最优路径研究不足问题,提出一种基于累积搜索概率理论的移动搜索平台路径优化方法。基于传感器性能评价的重要标准之一的累积探测概率(cumulative detection probability,CDP),利用时序相关性模型,即(λ,σ)过程模型,构造单峰CDP计算公式。构建一组目标运动想定,利用贝叶斯后验概率修正目标想定轨迹概率和不同时刻下的CDP。以搜索完成时CDP最大以及CDP达到目标水平时间最短为多目标,在连续时间与连续空间中实现高效搜索,构建面向声纳搜索移动目标的路径优化模型,通过多目标遗传算法给出优化解。与随机搜索模式下的CDP结果比对可以发现,本文方法可以获得更高的CDP,比单目标优化结果所得到的搜索方案具有更高的效率。

联合多波束与侧扫声纳的海缆检测方法

为解决海底电缆路由检测中多波束扫测分辨率低、侧扫声纳(SSS)图像位置不准且存在误识别问题,提出了一种联合多波束与侧扫声纳的海底电缆路由自动检测方法,利用多波束图像高精度的位置信息纠正侧扫声纳图像,并顾及侧扫声纳图像海缆成像特点进行线状特征增强与边缘提取,实现海底电缆目标的自动、准确检测。实验结果表明,该方法能实现非掩埋海缆探测的准确探测,为海底电缆检测提供了多源测量信息融合的新思路,丰富了海底电缆检测的技术方法。

基于前视声呐图像的AUV目标识别与跟踪

声呐图像由于水体不均匀、边界不规则以及声呐设备本身性能的限制,导致图像噪声明显、亮度不均、分辨率低,使得水下AUV装备在使用前视声呐进行水下目标检测时难度较大。针对该问题,基于m750d声呐探测获得的AUV声呐数据,进行了数据提取、高斯滤波处理、扇形映射处理,并采用Jet映射对声呐灰度图像进行了伪彩色映射提高数据标注速度和精度,制作获得了4组2 500张声呐图像的AUV目标检测数据集;采用YOLOv4-tiny目标检测算法开展AUV目标检测研究,研究结果表明该方法在该数据集上表现优秀,mAP@0.50达到94.17%,FPS在22帧左右,说明该轻量级网络在水下AUV目标识别与跟踪应用上具有较好的应用价值。

基于改进SSD的合成孔径声纳图像感兴趣小目标检测方法

轻量化目标检测模型SSD-MV3(Single Shot Detection-MobileNet V3)因输入图像尺寸限制无法直接检测高分辨率大尺寸合成孔径声纳(Synthetic Aperture Sonar,SAS)图像中感兴趣小目标.为此,本文提出了一种新的目标检测方法HRSSD(High Resolution Single Shot Detection),该方法通过冗余切割确保SSD-MV3输入图像尺寸的规范以及感兴趣小目标的完整,并利用二次非极大值抑制保证检测结果的唯一.此外,提出了一种尺度、空间和通道注意力机制联合的特征提取模块,并利用该模块重新设计了SSD-MV3的基础网络和附加特征提取网络,记作SSD-MV3P(Single Shot Detection-MobileNet V3 Pro),使得SSD-MV3P能更有效的感知感兴趣小目标特征信息.实验结果表明,在感兴趣小目标检测数据集SST(Sonar Small Targets)上,SSD-MV3P的平均检测精度(mean Average Precision,mAP)比SSD-MV3提升4.39%.HRSSD实现了高分辨率大尺寸SAS图像感兴趣小目标的检测,并且保证了同一位置上检测结果的完整性和唯一性.

基于ZYNQ的轻量化YOLOv5声呐图像目标检测算法及实现

针对声呐图像存在的模糊、样本量不足的现象,本文提出一种基于YOLOv5的声呐图像目标检测改进算法。利用几何滤波、垂直翻转等方法,对声呐图像数据集进行数据增强。添加融合注意力机制模块,使算法更好地关注声呐图像小目标的特征。同时,针对目前大多数目标检测算法运行在云端,无法做到实时性声呐图像检测的问题,本文利用替换轻量级网络和NCNN边端移植技术,同时在颈部网络中采用GSConv模块,将算法成功移植到ZYNQ平台,实现声呐图像的嵌入式端实时检测。实验表明,本文提出的算法在降低了56%参数量的同时,在map50和map50-95上分别提高2.2%和2.5%。改进后的算法性能提升明显,证明所提出的方法在轻量化声呐图像目标检测任务上具有一定的可行性与有效性。

ROV水下障碍物检测和避障技术的应用综述

全面回顾了远程操作车(ROV)在水下障碍物检测和避障技术方面的技术进展。研究集中于声呐系统、光学系统及其与机器学习和人工智能算法的结合,分析了这些技术如何提高水下作业的自主性、效率和安全性。尽管声纳和光学系统在环境适应性和障碍物检测精度方面已取得显著成果,但动态障碍物实时识别和复杂环境适应性的挑战仍待克服。此外,探讨了机器学习和人工智能技术在增强ROV自主避障能力方面的潜力和挑战,指出了这些技术在未来ROV操作中的重要性。该研究为深海探索和海洋科学提供了新的理论视角和应用实践。

基于非负矩阵分解的双曲调频信号目标回波增强

双曲调频信号是主动声呐中常用的多普勒不敏感信号。在实际使用中,由于信道变化和混响影响,目标回波较弱。针对水下航行器主动声呐探测性能下降问题,本文提出基于非负矩阵分解的双曲调频信号目标回波增强方法。该方法对目标回波和干扰信号的基矩阵分别定义,将发射信号的时频图作为目标回波的频率信息矩阵,对目标回波的时间信息矩阵做出约束,迭代处理完成目标回波的增强,获得较高增益的匹配回波检测结果。数值仿真和海上试验结果表明:本文方法与传统匹配滤波方法相比,其峰均比值提高了2 dB以上,提高了回波检测结果。

基于谱残差法的主动声呐疑似目标自动拾取

针对主动声呐干扰遍布的B显画面上目标亮点难以拾取的问题,本文提出一种基于谱残差法视觉显著性处理的疑似目标自动拾取方法。在主动声呐图像的概念之下,研讨了自然图像中视觉显著体的发现与主动声呐图像中疑似目标拾取的相似机理,依此进而说明了本文方法的机理、核心框架以及关键处理环节,包括预处理声呐图像、生成显著图、估算拾取阈值以及标定疑似目标等。海试数据处理结果表明:相较于基于局部相对能量分析的对照方法,本文方法对已知目标具有更低的漏报率,同时亮点拾取数量减少约40%,且处理耗时减少约37%。

基于卷积神经网络的声呐图像水下目标检测综述

声呐探测技术作为水下探测的主要手段之一,在海洋环境中具有广泛的应用,随着卷积神经网络(CNN)在计算机视觉领域表现出卓越的性能,该技术越来越受到研究人员的重视。文章综述了卷积神经网络在声呐图像水下目标检测中的应用与发展,重点讨论基于CNN的典型水下目标检测方法,包括基于候选区域与回归的检测算法在声呐图像上的改进与应用,并分析各类网络模型在处理声呐图像特有问题上的创新策略,如小样本检测、小目标检测及CNN与传统算法融合等。最后总结了当前基于CNN的水下目标检测面临的挑战,并预测该领域技术发展趋势。

基于随机共振的水下目标微弱回波信号增强检测

针对低信噪比下主动声呐探测水下目标的微弱回波信号检测难题,本文提出了一种基于随机共振的微弱回波信号增强检测方法。该方法通过建立频移变尺度随机共振系统,与输入高频微弱回波信号、随机噪声相匹配,实现微弱回波信号的共振增强;进而利用乘积谱与谱峭度构建目标回波亮点增强检测指标,有效提升微弱回波亮点的能量集中程度,抑制杂波及虚假成分干扰。仿真结果表明:在低信噪比下,随机共振系统输出信噪比增益可达5 dB以上,微弱目标回波亮点检测结果更加直观便捷,杂波干扰得到大幅抑制。海试数据处理结果进一步验证了所提方法的有效性和可靠性。

用于水下声呐目标检测的弱特征共焦通道调控方法

声呐图像视觉检测是复杂水域资源勘探和水下异物目标探测领域的重要技术之一。针对声呐图像中小目标存在的特征微弱和背景信息干扰问题,本文提出弱特征共焦通道调控水下声呐目标检测算法。为了提高模型对弱小目标的信息捕获和表征能力,设计弱小目标特征激活策略,并引入先验框尺度校准机制匹配底层语义特征检测分支,以提高小目标检测精度。应用全局信息聚合模块深入挖掘弱小目标的全局特征,避免冗余信息覆盖小目标微弱关键特征。为解决传统空间金字塔池化易忽视通道信息的问题,提出共焦通道调控池化模块,保留有效通道域小目标信息并克服复杂背景信息干扰。实验结果表明,本文所提模型在水下声呐数据集的9类弱小目标识别的平均检测精度达83.3%,相较基准提高了5.5%,其中铁桶、人体模型和立方体检测精度得到显著提高,分别提高24%、8.6%和7.3%,有效改善水下复杂环境中弱小目标漏检和误检问题。