[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122894-en":3,"doc-seo-122894-105":29,"detail-sidebar-cat-0-en-105":89},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":20,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":26,"seo_description":14,"update_tm":27,"read_time":28},122894,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","Data science for understanding physics - Modelling detectability of ship wake components using machine learning","Study models how detectability of individual ship-wake components in satellite SAR imagery depends on nine influencing parameters. Using SAR data from four missions with 4000 wake samples, wake length is manually measured and normalized into a Detectable Length Metric (DLM). A Support Vector Regression model maps DLM to environmental parameters, SAR acquisition settings, and ship properties, producing hyperplane and heatmap visualizations over the nine-dimensional feature space. Sensor-wise model comparisons show differing detection suitability, with results consistent with existing wake-recognition literature.","Data science for understanding physics –  \nmodelling detectability of ship wake components using machine learning  \nBjörn Tings*, Karl Kortum  \nGerman Aerospace Center (DLR), Remote Sensing Technology Institute Bremen  \n* [bjoern.tings@dlr.de](bjoern.tings@dlr.de)  \nFigure 1 (bottom right): Schematic representation of components of ship wakes detectable in Synthetic Aperture Radar (SAR) imagery.  \nFigure 1 (top left): Motivation for this study is provided by a campaign in 2014 with Polish Coast Guard and Federal Police Sea. The wakes generated by vessels with movement parallel to Range direction (horizontal) are worse detectable than wake generated by vessels with movement parallel to Azimuth direction (vertical).  \nIntroduction & Motivation  \nShip wakes consists of multiple components, which are detectable in acquisitions from satelliteborne Synthetic Aperture Radar (SAR) sensors. Their detectability varies in dependency to influencing parameters, which can be categorized into the types: environmental parameters, image acquisition settings and ship properties. An example of varying detectability of ship wakes and a schematic representation of detectable wake components are presented in Figure 1.  \nIn this study, machine learning (i.e. Support Vector Regression (SVR)) is used to model the dependency between nine influencing parameters and the detectability of individual wake components.  \nData  \nSAR data from four satellite SAR missions is used. The SAR datasets with in total 4000 wake samples are summarized in Table 1. The nine influencing parameters affecting the detectability of each of the wake samples are described in Table 2. The length of each wake component is measured, based on a manual retracing of the wake component’s outlines. The plotsin Figure 2 a) provide an exemplary view into the feature space spanned by two of the influencing parameters with length of awake component, which is indicating the respective wake component’s detectability.  \nFigure 2 a): Feature space with all wake samples of TSX  \ndataset for two influencing parameters and wake length  \nTable 1: Summary of SAR datasets for SAR missions TerraSAR-X (TSX), CosmoSkymed (CSK), Sentinel-1 (S1) and RADARSAT-2 (RS2)  \n\n|  | TSX | CSK | S1 | RS2 |\n| --- | --- | --- | --- | --- |\n| Frequency band | X | X | C | C |\n| Acquisitions modes | SL, SM | HI | IW | MF, F, S |\n| Amount of products | 1097 | 11 | 31 | 53 |\n| Amount of wake samples | 2881 | 94 | 618 | 407 |\n\nTable 2: List of nine influencing parameters investigated with respect to their influence on the detectability of individual wake components.  \n\n| Nr i, | Influencing Parameter Name\u003Cbr>(􀢞 􀢏) | Description |\n| --- | --- | --- |\n| 1 | AIS-VesselVelocity (􀝔1 ) | Velocity of the vessel derived from AIS |\n| 2 | AIS-Length (􀝔2 ) | Length of the vessel derived from AIS |\n| 3 | AIS-CoG (􀝔3 ) | Course over ground (CoG) derived from AIS relative to the radar looking direction |\n| 4 | Incidence-Angle (􀝔4 ) | Incidence angle of the radar |\n| 5 | SAR-Wind-Speed (􀝔5 ) | Wind speed estimated from SAR backscatter of ocean background |\n| 6 | SAR-SignificantWave-Height (􀝔6 ) | Significant wave height estimated from SAR backscatter of ocean background |\n| 7 | SAR-Wave-Length (􀝔7 ) | Wavelength estimated from from SAR backscatter of ocean background |\n| 8 | AIS-CoG-SARWave-Direction (􀝔8 ) | Absolute angular difference between AISCoG and wave direction estimated from SAR backscatter of ocean background |\n| 9 | AIS-CoG-WRFWind-Direction (􀝔9 ) | Absolute angular difference between AISCoG and wind direction estimated by the Weather Research and Forecasting Model (WRF) |\n\nFigure 2 b): Gray hyperplane visualizing detectability model for turbulent wakes based on displayed wake samples  \nMethod  \nIn order to investigate the detectability of a wake component 􀝓 , a normalized figure of merit for wake detectability is required. In this study, the indicator of detectability, i.e. wake component length, is normalized between a minim","cbCaisDNwETJP43o","https://ap.wps.com/l/cbCaisDNwETJP43o","pdf",1195325,1,"English","en",105,"# Introduction & Motivation\n# Data\n## SAR datasets and wake samples\n## Influencing parameters\n# Method\n## Detectable Length Metric (DLM)\n## Support Vector Regression (SVR) mapping\n## Heatmaps and hyperplane visualization\n# Conclusion","[{\"question\":\"What does the study model and which measure represents wake detectability?\",\"answer\":\"The study models the relationship between influencing parameters and detectability of individual ship-wake components in SAR imagery. Wake component length is normalized into the Detectable Length Metric (DLM) to represent detectability.\"},{\"question\":\"How is the machine-learning model constructed in this work?\",\"answer\":\"A Support Vector Regression (SVR) model is trained to map DLM to nine influencing parameters. The feature space is sampled using discretized parameter tuples, and DLM values are computed for detection conditions.\"},{\"question\":\"Which types of parameters affect wake detectability according to the document?\",\"answer\":\"Detectability depends on environmental parameters, SAR image acquisition settings, and ship properties. The nine investigated parameters include vessel and course information from AIS, incidence angle, and ocean-wave and direction-related quantities.\"}]","Data science for understanding physics - Modelling detectability of ship wake components using machine learning | PDF",1785813542,3,{"code":4,"msg":30,"data":31},"ok",{"site_id":23,"language":22,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":84,"head_meta":86,"extra_data":88,"updated_unix":27},"data-science-for-understanding-physics-modelling-detectability-of-ship-wake-components-using-machine-learning","",{"@graph":35,"@context":83},[36,52,66],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,49],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":28},"https://docshare.wps.com/document/research-report/",{"item":50,"name":13,"@type":42,"position":51},"https://docshare.wps.com/document/data-science-for-understanding-physics-modelling-detectability-of-ship-wake-components-using-machine-learning/122894/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":22,"description":14,"dateModified":60,"datePublished":60,"encodingFormat":59,"isAccessibleForFree":61,"interactionStatistic":62},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":63,"interactionType":64,"userInteractionCount":4},"InteractionCounter",{"@type":65},"ViewAction",{"@type":67,"mainEntity":68},"FAQPage",[69,75,79],{"name":70,"@type":71,"acceptedAnswer":72},"What does the study model and which measure represents wake detectability?","Question",{"text":73,"@type":74},"The study models the relationship between influencing parameters and detectability of individual ship-wake components in SAR imagery. Wake component length is normalized into the Detectable Length Metric (DLM) to represent detectability.","Answer",{"name":76,"@type":71,"acceptedAnswer":77},"How is the machine-learning model constructed in this work?",{"text":78,"@type":74},"A Support Vector Regression (SVR) model is trained to map DLM to nine influencing parameters. The feature space is sampled using discretized parameter tuples, and DLM values are computed for detection conditions.",{"name":80,"@type":71,"acceptedAnswer":81},"Which types of parameters affect wake detectability according to the document?",{"text":82,"@type":74},"Detectability depends on environmental parameters, SAR image acquisition settings, and ship properties. The nine investigated parameters include vessel and course information from AIS, incidence angle, and ocean-wave and direction-related quantities.","https://schema.org",{"og:url":50,"og:type":85,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":87,"canonical":50},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":90},[91,95,99,103,108,113,118,121,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":92,"show_sort_weight":93,"slug":94},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":96,"show_sort_weight":97,"slug":98},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":45,"category_name":100,"show_sort_weight":101,"slug":102},"Exam",70,"exam",{"id":104,"doc_module":4,"doc_module_name":45,"category_name":105,"show_sort_weight":106,"slug":107},5,"Comic",60,"comic",{"id":109,"doc_module":4,"doc_module_name":45,"category_name":110,"show_sort_weight":111,"slug":112},6,"Technology",50,"technology",{"id":114,"doc_module":4,"doc_module_name":45,"category_name":115,"show_sort_weight":116,"slug":117},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":119,"slug":120},30,"research-report",{"id":122,"doc_module":4,"doc_module_name":45,"category_name":123,"show_sort_weight":124,"slug":125},9,"Religion & Spirituality",20,"religion-spirituality",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":124,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":104,"slug":136},19,"General","general"]