基于深度学习的海洋平台基础环境监测方法

doi:10.13474/j.cnki.11-2246.2025.0603

测绘通报 ›› 2025, Vol. 0 ›› Issue (6): 12-17.doi: 10.13474/j.cnki.11-2246.2025.0603

基于深度学习的海洋平台基础环境监测方法

张超^1,2, 熊春宝¹, 连思达^1,3

1. 天津大学建筑工程学院, 天津 300072;
2. 中海石油技术检测有限公司, 天津 300457;
3. 河北工程大学土木工程学院, 河北邯郸 056009

收稿日期:2024-11-13 发布日期:2025-07-04
作者简介:张超(1990—),男,工程师,主要从事测绘工程的生产。E-mail:zhangchao42@cnooc.com.cn
基金资助:
国家自然科学基金面上项目(61971037)

Deep learning-based method for foundation environmental monitoring of offshore platform

ZHANG Chao^1,2, XIONG Chunbao¹, LIAN Sida^1,3

1. School of Civil Engineering, Tianjin University, Tianjin 300072, China;
2. CNOOC Technical Testing Co., Ltd., Tianjin 300457, China;
3. School of Civil Engineering, Hebei University of Engineering, Handan 056009, China

Received:2024-11-13 Published:2025-07-04

摘要/Abstract

摘要： 针对海洋平台基础海底环境监测人工工作效率低,时间成本高,依赖具有先验知识的技术人员的主观判断等问题,本文提出了一种基于深度学习的智能化海洋平台基础环境监测算法。该算法能够有效避免人工主观因素对图像判读的干扰,实现全天候、自动化、实时的海洋平台基础环境灾害初步预警。首先,通过试验从目前基于深度学习(DL)的主流计算机视觉(CV)算法中筛选出最适合海底三维实时声呐图像分类的基础网络结构。然后,对试验选出的VGG-11算法,使用通道优先卷积注意力(CPCA)模块改进,并通过Grad-CAM热力图验证了CPCA-VGG算法的有效性。试验结果表明,CPCA-VGG算法对海洋平台基础各类环境分类评价指标分别达到了Acc_Top-1 97.35%、Acc_Top-5 100.00%,平均精确率和平均召回率分别为98.62%、98.44%。该算法能较好地满足对海洋平台基础各类环境的实时监测,以及对灾害进行初步预警的实际工程需要。

关键词: 结构健康监测, 海洋平台, 海底环境监测, 计算机视觉, 深度学习, 三维实时声呐图像

Abstract: Structural health monitoring (SHM) of offshore platform and ancillary pipelines pose challenges, including low efficiency of manual labor, high time costs, and reliance on subjectivity of technicians with existing knowledge. To address these issues, an intelligent, deep learning (DL)-based algorithm for monitoring the foundation subsea environment of offshore platform was proposed. This approach can prevent manual subjective factors from affecting image interpretation and enable early detection of foundation environmental disasters using an all-weather automated real-time offshore platform. Firstly, experiments were conducted to determine the most appropriate basic network structure for real-time classification of subsea 3D sonar images using computer vision(CV) algorithms based on DL that are presently in use. Secondly, the channel-priority convolutional attention (CPCA) module was employed for the improvement of the experimentally selected VGG-11 algorithm, and the effectiveness of the CPCA-VGG algorithm was verified by the Grad-CAM algorithm. The experimental results demonstrate that the CPCA-VGG algorithm assessment criteria achieve: Acc_Top-1 97.35%,Acc_Top-5 100.00%, mean precision and mean recall is 98.62% and 98.44%, when it was applied in offshore platform foundation environment classification. This algorithm can better meet the practical engineering needs of real-time monitoring of various environments based on offshore platforms and preliminary early warning of disasters.

Key words: structural health monitoring, offshore platform, subsea environment monitoring, computer vision, deep learning, 3D real-time sonar image

中图分类号:

P229

张超, 熊春宝, 连思达. 基于深度学习的海洋平台基础环境监测方法[J]. 测绘通报, 2025, 0(6): 12-17.

ZHANG Chao, XIONG Chunbao, LIAN Sida. Deep learning-based method for foundation environmental monitoring of offshore platform[J]. Bulletin of Surveying and Mapping, 2025, 0(6): 12-17.

参考文献

[1] JIANG Chunmeng,WAN Lei,SUN Yushan. Design of motion control system of pipeline detection AUV[J]. Journal of Central South University,2017,24(3):637-646.
[2] ZHANG Xiaolong,DONG Yang,XU Xiang,et al. A new strategy for tube leakage and blockage detection using bubble motion-based solid-liquid triboelectric sensor[J]. Science China Technological Sciences,2022,65(2):282-292.
[3] SHELTAMI T R,BALA A,SHAKSHUKI E M. Wireless sensor networks for leak detection in pipelines:a survey[J]. Journal of Ambient Intelligence and Humanized Computing,2016,7(3):347-356.
[4] ACOSTA G G,VILLAR S A. Accumulated CA-CFAR process in 2D for online object detection from sidescan sonar data[J]. IEEE Journal of Oceanic Engineering,2015,40(3):558-569.
[5] KONG Wanzeng,HONG Jichen,JIA Mingyang,et al. YOLOv3-DPFIN:a dual-path feature fusion neural network for robust real-time sonar target detection[J]. IEEE Sensors Journal,2020,20(7):3745-3756.
[6] NHAT-DUC H,NGUYEN Q L,TRAN V D. Automatic recognition of asphalt pavement cracks using metaheuristic optimized edge detection algorithms and convolution neural network[J]. Automation in Construction,2018,94:203-213.
[7] GOPALAKRISHNAN K,KHAITAN S K,CHOUDHARY A,et al. Deep convolutional neural networks with transfer learning for computer vision-based data-driven pavement distress detection[J]. Construction and Building Materials,2017,157:322-330.
[8] KIM B,CHO S. Image-based concrete crack assessment using mask and region-based convolutional neural network[J]. Structural Control and Health Monitoring,2019:e2381.
[9] JANG K,KIM N,AN Y K. Deep learning-based autonomous concrete crack evaluation through hybrid image scanning[J]. Structural Health Monitoring,2019,18(5/6):1722-1737.
[10] ZHU Jinsong,ZHANG Chi,QI Haidong,et al. Vision-based defects detection for bridges using transfer learning and convolutional neural networks[J]. Structure and Infrastructure Engineering,2020,16(7):1037-1049.
[11] ALIPOUR M,HARRIS D K. Increasing the robustness of material-specific deep learning models for crack detection across different materials[J]. Engineering Structures,2020,206:110157.
[12] XU Yang,BAO Yuequan,CHEN Jiahui,et al. Surface fatigue crack identification in steel box girder of bridges by a deep fusion convolutional neural network based on consumer-grade camera images[J]. Structural Health Monitoring,2019,18(3):653-674.
[13] MA Duo,LIU Jianhua,FANG Hongyuan,et al. A multi-defect detection system for sewer pipelines based on StyleGAN-SDM and fusion CNN[J]. Construction and Building Materials,2021,312:125385.
[14] HUANG H,CHEN Z,ZOU Y,et al. Channel prior convolutional attention for medical image segmentation[J]. Computers in Biology and Medicine,2024,178:108784.
[15] SELVARAJU R R,COGSWELL M,DAS A,et al. Grad-CAM:visual explanations from deep networks via gradient-based localization[J]. International Journal of Computer Vision,2020,128(2):336-359.

基于深度学习的海洋平台基础环境监测方法

Deep learning-based method for foundation environmental monitoring of offshore platform

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	闫白羽, 翟国君, 边少锋. 基于白化着色的沉船和失事飞机风格迁移方法[J]. 测绘通报, 2025, 0(6): 73-77.
[2]	陈登峰, 何拓航, 杨小燕, 刘世鹏, 孟屯良. 路基病害雷达图像多视图融合识别方法[J]. 测绘通报, 2025, 0(5): 66-73.
[3]	赵辰瑞, 连增增, 田亚林, 贺刘辉, 陈浩, 王鹏辉, 王孟奇. BiLSTM-Chan算法在超宽带室内定位中的应用[J]. 测绘通报, 2025, 0(5): 100-105.
[4]	朱松, 陈安泰, 华远盛, 姜文宇, 元鹏鹏, 朱家松. 面向漂流胶囊机器人图像的管道缺陷检测[J]. 测绘通报, 2025, 0(4): 34-38.
[5]	马于博, 张爱国, 王浩宇, 刘帅琪, 靳镜宇, 沈占锋, 李均力. 基于U-ConvHDNet模型的戈壁砾幕层提取[J]. 测绘通报, 2025, 0(4): 68-74.
[6]	李建, 王健, 王雷, 李敏, 杨立克, 赵艺龙. 双重注意力机制的电力走廊点云语义分割[J]. 测绘通报, 2025, 0(4): 127-133.
[7]	谢巴图, 胡佳睿, 潘俊. 基于双路全局信息优化网络的遥感影像海陆分割算法[J]. 测绘通报, 2025, 0(3): 52-58,86.
[8]	张璇, 杨本勇, 文武, 邓维熙, 周何帆. 改进的DeepLabV3+模型用于四川县域大小春作物识别[J]. 测绘通报, 2025, 0(3): 133-137,167.
[9]	苟长龙, 庞敏, 杨扬. 改进的U-Net卷积网络在遥感影像地物分类中的应用[J]. 测绘通报, 2025, 0(3): 150-155.
[10]	马海志. 基于深度学习的隧道超高分辨率图像病害检测框架[J]. 测绘通报, 2025, 0(3): 105-110.
[11]	宋培焱, 叶勤, 曾亮, 罗俊奇, 张硕, 尹长江. 面向密度明显差异点云的室内场景三维Mesh模型POCO重建方法[J]. 测绘通报, 2025, 0(2): 7-12,47.
[12]	石佳俊, 臧玉府, 肖雄武, 张莹滢. 非刚性匹配的激光点云与多光谱影像深度融合[J]. 测绘通报, 2025, 0(2): 41-47.
[13]	马浩然, 王金华. 多维特征学习与模型融合的地理命名实体识别[J]. 测绘通报, 2024, 0(9): 129-134.
[14]	石一剑, 谭海, 钟旭辉. 改进YOLOv5的国产光学影像辐射异常检测方法[J]. 测绘通报, 2024, 0(8): 60-65,72.
[15]	武广臣, 刘艳. 无人机影像GrabCut路面裂缝识别[J]. 测绘通报, 2024, 0(8): 90-95.