Huixiang Huang , Qiaoling Yan , Yang Yang , Yu Hu , Shaohua Wang , Qirui Yuan , Xiao Li , Qiang Mei
{"title":"基于深度学习的港口设施空间分类模型和港口管理能源储备预测--以宁波为例","authors":"Huixiang Huang , Qiaoling Yan , Yang Yang , Yu Hu , Shaohua Wang , Qirui Yuan , Xiao Li , Qiang Mei","doi":"10.1016/j.ocecoaman.2024.107413","DOIUrl":null,"url":null,"abstract":"<div><div>Port facilities and energy storage capacity significantly affect maritime logistics efficiency and supply chain security, necessitating accurate and timely port facility information. However, unavailable real-time open port data complicate effective quantitative evaluations of port development along the 21st-Century Maritime Silk Road. This research addresses these issues by combining deep learning with remote sensing, using image data from key ports. A method was proposed to classify port facilities and energy reserve information spatially. A multi-classification framework using U-Net semantic segmentation was developed to segment key facilities in remote sensing data sets. The You Only Look Once v8 (YOLOv8) model was applied to locate oil tanks within Ningbo Port. The actual roof area of oil tanks was then extracted using a deep learning model, facilitating statistical analysis and comparative studies with other major oil and gas ports. Additionally, the real-time remote sensing image index calculated oil tank heights based on shadow lengths for capacity measurement of floating roof tanks. Experimental results showed a pixel accuracy of 90% and an intersection over union of 84% for oil tank region extraction, with an oil tank recognition model achieving a mean average precision of 98.9%. Compared with traditional Hough transform methods, the average absolute error, average relative error, and standard deviation for tank roof area calculations were reduced by 229.18 m<sup>2</sup>, 4.6%, and 81%, respectively. This framework effectively determined the number of oil tanks in various ports, enabling real-time reserve detection and providing a data foundation for energy port management and resilience research.</div></div>","PeriodicalId":54698,"journal":{"name":"Ocean & Coastal Management","volume":"258 ","pages":"Article 107413"},"PeriodicalIF":4.8000,"publicationDate":"2024-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Spatial classification model of port facilities and energy reserve prediction based on deep learning for port management―A case study of Ningbo\",\"authors\":\"Huixiang Huang , Qiaoling Yan , Yang Yang , Yu Hu , Shaohua Wang , Qirui Yuan , Xiao Li , Qiang Mei\",\"doi\":\"10.1016/j.ocecoaman.2024.107413\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Port facilities and energy storage capacity significantly affect maritime logistics efficiency and supply chain security, necessitating accurate and timely port facility information. However, unavailable real-time open port data complicate effective quantitative evaluations of port development along the 21st-Century Maritime Silk Road. This research addresses these issues by combining deep learning with remote sensing, using image data from key ports. A method was proposed to classify port facilities and energy reserve information spatially. A multi-classification framework using U-Net semantic segmentation was developed to segment key facilities in remote sensing data sets. The You Only Look Once v8 (YOLOv8) model was applied to locate oil tanks within Ningbo Port. The actual roof area of oil tanks was then extracted using a deep learning model, facilitating statistical analysis and comparative studies with other major oil and gas ports. Additionally, the real-time remote sensing image index calculated oil tank heights based on shadow lengths for capacity measurement of floating roof tanks. Experimental results showed a pixel accuracy of 90% and an intersection over union of 84% for oil tank region extraction, with an oil tank recognition model achieving a mean average precision of 98.9%. Compared with traditional Hough transform methods, the average absolute error, average relative error, and standard deviation for tank roof area calculations were reduced by 229.18 m<sup>2</sup>, 4.6%, and 81%, respectively. This framework effectively determined the number of oil tanks in various ports, enabling real-time reserve detection and providing a data foundation for energy port management and resilience research.</div></div>\",\"PeriodicalId\":54698,\"journal\":{\"name\":\"Ocean & Coastal Management\",\"volume\":\"258 \",\"pages\":\"Article 107413\"},\"PeriodicalIF\":4.8000,\"publicationDate\":\"2024-10-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Ocean & Coastal Management\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0964569124003983\",\"RegionNum\":2,\"RegionCategory\":\"环境科学与生态学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"OCEANOGRAPHY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Ocean & Coastal Management","FirstCategoryId":"93","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0964569124003983","RegionNum":2,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"OCEANOGRAPHY","Score":null,"Total":0}
引用次数: 0
摘要
港口设施和储能能力对海运物流效率和供应链安全有着重大影响,因此需要准确及时的港口设施信息。然而,由于无法获得实时开放的港口数据,对 21 世纪海上丝绸之路沿线港口发展的有效定量评估变得更加复杂。本研究利用主要港口的图像数据,将深度学习与遥感技术结合起来,解决了这些问题。研究提出了一种对港口设施和能源储备信息进行空间分类的方法。利用 U-Net 语义分割技术开发了一个多分类框架,用于分割遥感数据集中的关键设施。应用YOLOv8(You Only Look Once v8)模型对宁波港内的油罐进行定位。然后利用深度学习模型提取油罐的实际顶盖面积,以便进行统计分析和与其他主要油气港口的比较研究。此外,实时遥感图像指数根据阴影长度计算油罐高度,用于浮顶油罐的容量测量。实验结果表明,油罐区域提取的像素精确度为 90%,交集超过联合的精确度为 84%,油罐识别模型的平均精确度达到 98.9%。与传统的 Hough 变换方法相比,油罐顶面积计算的平均绝对误差、平均相对误差和标准偏差分别减少了 229.18 平方米、4.6% 和 81%。该框架有效确定了不同港口的油罐数量,实现了实时储备检测,为能源港口管理和抗灾研究提供了数据基础。
Spatial classification model of port facilities and energy reserve prediction based on deep learning for port management―A case study of Ningbo
Port facilities and energy storage capacity significantly affect maritime logistics efficiency and supply chain security, necessitating accurate and timely port facility information. However, unavailable real-time open port data complicate effective quantitative evaluations of port development along the 21st-Century Maritime Silk Road. This research addresses these issues by combining deep learning with remote sensing, using image data from key ports. A method was proposed to classify port facilities and energy reserve information spatially. A multi-classification framework using U-Net semantic segmentation was developed to segment key facilities in remote sensing data sets. The You Only Look Once v8 (YOLOv8) model was applied to locate oil tanks within Ningbo Port. The actual roof area of oil tanks was then extracted using a deep learning model, facilitating statistical analysis and comparative studies with other major oil and gas ports. Additionally, the real-time remote sensing image index calculated oil tank heights based on shadow lengths for capacity measurement of floating roof tanks. Experimental results showed a pixel accuracy of 90% and an intersection over union of 84% for oil tank region extraction, with an oil tank recognition model achieving a mean average precision of 98.9%. Compared with traditional Hough transform methods, the average absolute error, average relative error, and standard deviation for tank roof area calculations were reduced by 229.18 m2, 4.6%, and 81%, respectively. This framework effectively determined the number of oil tanks in various ports, enabling real-time reserve detection and providing a data foundation for energy port management and resilience research.
期刊介绍:
Ocean & Coastal Management is the leading international journal dedicated to the study of all aspects of ocean and coastal management from the global to local levels.
We publish rigorously peer-reviewed manuscripts from all disciplines, and inter-/trans-disciplinary and co-designed research, but all submissions must make clear the relevance to management and/or governance issues relevant to the sustainable development and conservation of oceans and coasts.
Comparative studies (from sub-national to trans-national cases, and other management / policy arenas) are encouraged, as are studies that critically assess current management practices and governance approaches. Submissions involving robust analysis, development of theory, and improvement of management practice are especially welcome.