[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126581-en":3,"doc-seo-126581-105":31,"detail-sidebar-cat-0-en-105":92},{"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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},126581,687207017582,"Himbo","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Automatic Roof Type Classification Through Machine Learning for Regional Wind Risk Assessment","Roof type is a key building characteristic for wind vulnerability modeling, yet it is often absent from publicly available datasets. A machine-learning framework is developed to generate high-resolution roof-type data from building-level satellite imagery. A convolutional neural network classifies roof types with an F1 score of 0.96 on 1,000 test buildings, then predicts roof types for 161,772 single-family houses in two U.S. counties. Roof-type distributions are analyzed at city and census-tract scales, and missing values are restored using imputation algorithms leveraging building attributes and neighborhood-level roof characteristics.","arXiv :2305 . 17315v1 [ cs .LG] 27 May 2023  \nAutomatic Roof Type Classification Through Machine Learning for  \nRegional Wind Risk Assessment  \nShuochuan Menga,∗, Mohammad Hesam Soleimani-Babakamalia , Ertugrul Taciroglua  \na Department of Civil and Environmental Engineering, University of California, Los Angeles, CA, United States  \nAbstract  \nRoof type is one of the most critical building characteristics for wind vulnerability modeling. It is also the most frequently missing building feature from publicly available databases. An automatic roof classification framework is developed herein to generate high-resolution roof-type data using machine learning. A Convolutional Neural Network (CNN) was trained to classify roof types using building-level satellite images. The model achieved the F1 score of 0.96 on predicting roof types for 1,000 test buildings. The CNN model was then used to predict roof types for 161,772 single-family houses in New Hanover County, NC, and Miami-Dade County, FL. The distribution of roof type in city and census tract scales was presented. A high variance was observed in the dominant roof type among census tracts. To improve the completeness of the roof-type data, imputation algorithms were developed to populate missing roof data due to low-quality images, using critical building attributes and neighborhood-level roof characteristics.  \nKeywords: Machine learning; Deep learning; Roof type classification; Data Imputation;  \nHurricanes.  \n1. Introduction  \nRoof type is one of the most critical building characteristics for wind vulnerability modeling of residential buildings. In hurricane-prone regions of the United States, gable and hip roofs are the dominant roof types for single-family houses [1, 2] . It was observed in wind tunnels tests [3–5] and post-disaster survey [6, 7] that gable and hip roofs experience distinctive wind pressures and damage frequency under high winds. Therefore, the quality of roof-type data is crucial for the accuracy  \n∗ Corresponding author  \nEmail address: [vessel@ucla.edu](vessel@ucla.edu) (Shuochuan Meng)  \nPreprint submitted to Advanced Engineering Informatics May 30, 2023  \nand reliability of regional wind risk assessments. Nevertheless, roof type is the also most frequently missing building feature from publicly available databases such as tax appraisers’ databases [1] .  \nSeveral studies have focused on populating roof-type data using machine learning-based methods, which can be summarized into two major types. One type of approach is predicting roof type using other building features [8, 9] with machine learning models. Pita et al. [8] applied Bayesian Belief Networks (BBN) and Classification and Regression Trees (CART) to impute missing rooftype data using building characteristics such as construction year and building value. Mohajeri et al. [9] used Support Vector Machine (SVM) to classify roof types using roof geometries obtained from LiDAR data. However, the prediction of roof types heavily relied on the availability of other building information. Moreover, building features, like building value, are correlated with building location, which deteriorates the performance of the model when applied to different areas [8] . The other approach is classifying roof types using remote sensing data with Convolutional Neural Networks (CNNs) through Representation Learning [10] . Representation learning extracts the features automatically through its learning process, without the need for hand-crafted features [11], thus offering higher generalizability than statistical and hand-crafted features which might not generalize to unseen data well [12] . Various CNNs were trained to predict roof type using neighborhood-level satellites image [13], single-building-level satellite images [14, 15], and the combination of satellite images and LiDAR data [16] . It was demonstrated that CNNs could classify roof types efficiently with high accuracy. Nevertheless, low-quality satellite ima","cbCailXh97gEzJvh","https://ap.wps.com/l/cbCailXh97gEzJvh","pdf",29522087,4,1,31,"English","en",105,"# Introduction\n## Roof type importance and data gaps\n## Related machine-learning approaches\n### Predicting roof types from other building features\n### Classifying from remote sensing data with CNNs\n## Wind vulnerability models and need for refined archetypes\n## Proposed framework overview","[{\"question\":\"Why is roof type crucial for regional wind risk assessment?\",\"answer\":\"Roof type strongly influences wind vulnerability modeling for residential buildings because different roof forms experience distinct wind pressures and damage patterns.\"},{\"question\":\"How does the proposed framework classify roof types?\",\"answer\":\"It trains a convolutional neural network to classify roof types from building-level satellite images, achieving an F1 score of 0.96 on test buildings.\"},{\"question\":\"How are missing roof-type labels handled when satellite images are low quality?\",\"answer\":\"Low-quality images are detected and removed, then missing roof-type data are imputed using building attributes and roof-type distribution characteristics from neighboring buildings.\"}]","Automatic Roof Type Classification Through Machine Learning for Regional Wind Risk Assessment | PDF",1785933460,78,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":29},"automatic-roof-type-classification-through-machine-learning-for-regional-wind-risk-assessment","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":20},"https://docshare.wps.com/document/automatic-roof-type-classification-through-machine-learning-for-regional-wind-risk-assessment/126581/",{"url":53,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-27","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"Why is roof type crucial for regional wind risk assessment?","Question",{"text":76,"@type":77},"Roof type strongly influences wind vulnerability modeling for residential buildings because different roof forms experience distinct wind pressures and damage patterns.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the proposed framework classify roof types?",{"text":81,"@type":77},"It trains a convolutional neural network to classify roof types from building-level satellite images, achieving an F1 score of 0.96 on test buildings.",{"name":83,"@type":74,"acceptedAnswer":84},"How are missing roof-type labels handled when satellite images are low quality?",{"text":85,"@type":77},"Low-quality images are detected and removed, then missing roof-type data are imputed using building attributes and roof-type distribution characteristics from neighboring buildings.","https://schema.org",{"og:url":53,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]