[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117488-en":3,"doc-seo-117488-105":30,"detail-sidebar-cat-0-en-105":91},{"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":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},117488,1099514067415,"Rowan","https://ap-avatar.wpscdn.com/avatar/100002539d78ffe74a7?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779092875211072502",8,"Research & Report","Kalman-Enhanced Streaming Machine Learning for Real-Time Land Use Classification in Satellite Imagery","Rapid escalation of data generation has pushed machine learning toward Streaming Machine Learning, designed to handle continuous, unbounded data streams. This thesis applies streaming learning to satellite imagery classification, where timely analysis supports environmental monitoring, disaster response, and urban planning. Instead of batch processing, the proposed approach enables real-time decisions and adaptation to changing data distributions. To manage high-dimensional satellite data, it introduces an optimized Kalman-SLDA pipeline using Kalman filtering with Streaming Linear Discriminant Analysis, paired with dimensionality reduction via Sparse Random Projection and streaming UMAP adaptation. Experiments show improved accuracy and efficiency under class imbalance and concept drift, enabling viable real-time geospatial processing.","Kalman-Enhanced Streaming Machine Learning for Real-Time Land Use Classification in Satellite Imagery  \nTesi di Laurea Magistrale in  \nMathematical Engineering-Ingegneria Matematica  \nAuthor: Nicola Francescon  \nStudent ID: 220697  \nAdvisor: Prof. Emanuele Della Valle  \nCo-advisors: Giacomo Ziffer  \nAcademic Year: 2023-24  \ni  \nAbstract  \nThe rapid escalation of data generation in recent years has oriented Machine Learning toward applications designed to manage continuous, unbounded data streams, in a field called Streaming Machine Learning. This thesis investigates the application of Streaming Machine Learning in the classification of satellite imagery, a domain where timely analysis of incoming data is essential for effective decision-making in areas such as environmental monitoring, disaster response, and urban planning. Unlike traditional batch processing, Streaming Machine Learning enables real-time decision-making and adaptation to changes in data distribution, ensuring models remain relevant as new data arrives. To address the computational challenges associated with high-dimensional satellite data, this research proposes an optimized pipeline incorporating Streaming Linear Discriminant Analysis with Kalman Filtering, namely Kalman-SLDA, for robust classification performance. Advanced dimensionality reduction techniques, including Sparse Random Projection anda novel adaptation of the UMAP algorithm for streaming environments, are employed to mitigate the computational load without compromising classification accuracy. Experimental results demonstrate that Kalman-SLDA outperforms traditional classifiers in terms of accuracy and efficiency, even in the presence of class imbalance and concept drift. Furthermore, the proposed pipeline leverages dimensionality reduction to improve computational speed, presenting a viable solution for the real-time processing demands of satellite data. The findings underscore the importance of developing efficient algorithms for streaming data and contribute to support rapid, data-driven decision-making in dynamic environments. This work provides a foundation for future research into adaptive, scalable classification pipelines for satellite imagery, with the potential to improve a wide range of applications in geospatial analysis.  \nKeywords: Streaming Machine Learning, Dimensionality reduction, Satellite Image Classification, Real-Time Processing, Kalman filtering  \nAbstract in lingua italiana  \nIl rapido aumento della generazione di dati negli ultimi anni ha condotto il Machine Learning verso applicazioni sempre più orientate a gestire flussi di dati continui e illimitati, un campo denominato Streaming Machine Learning. Questa tesi esplora l’applicazione dello Streaming Machine Learning nella classificazione delle immagini satellitari, un ambito in cui l’analisi tempestiva dei dati è essenziale per guidare le decisioni in contesti quali il monitoraggio ambientale, la risposta alle catastrofi e la pianificazione urbana. A differenza dei tradizionali approcci batch, lo Streaming Machine Learning consente di prendere decisioni in tempo reale e di adattarsi ad eventuali cambiamenti nella distribuzione dei dati, garantendo che i modelli rimangano accurati con l’afflusso di nuove informazioni. Per affrontare i problemi computazionali associati ai dati satellitari ad alta dimensionalità, questa ricerca propone una pipeline ottimizzata che implementa il classificatore SLDA integrato con il filtro di Kalman, chiamato Kalman-SLDA, mirando a garantire una classificazione robusta delle immagini. Tecniche avanzate di riduzione della dimensionalità, tra cui Sparse Random Projection e un innovativo adattamento dell’algoritmo UMAP per contesti streaming, sono impiegate per ridurre il carico computazionale mantenendo un’elevata precisione nella classificazione. I risultati sperimentali evidenzianoche Kalman-SLDA supera i classificatori tradizionali sia in termini di accuratezza che diefficienza, anche in pr","cbCaiq9Nbqlf6reL","https://ap.wps.com/l/cbCaiq9Nbqlf6reL","pdf",5291738,1,108,"English","en",105,"# 1 Introduction\n## 1.1 Contributions of the Thesis\n## 1.2 Thesis Outline\n# 2 Related Works\n## 2.1 Introduction to Streaming Machine Learning\n## 2.1.1 Concept Drift\n## 2.2 Streaming Machine Learning Classifiers\n## 2.2.1 Gaussian Naive Bayes\n## 2.2.2 Softmax Regression\n## 2.2.3 Streaming Linear Discriminant Analysis\n## 2.3 Convolutional Neural Networks\n## 2.3.1 MobileNet V3 Small\n## 2.3.2 ResNet 18\n## 2.4 Satellite Image Datasets\n## 2.4.1 Satellite Image Classification\n## 2.4.2 The Dataset: Functional Map of the World-Time","[{\"question\":\"What problem does the thesis address in satellite imagery classification?\",\"answer\":\"It addresses the need for timely, real-time classification while processing continuous and unbounded incoming satellite data streams.\"},{\"question\":\"How does Kalman-SLDA improve performance compared to traditional classifiers?\",\"answer\":\"The pipeline combines Streaming Linear Discriminant Analysis with Kalman filtering and dimensionality reduction, achieving higher accuracy and efficiency even with class imbalance and concept drift.\"},{\"question\":\"Why are dimensionality reduction techniques used in the proposed pipeline?\",\"answer\":\"They reduce the computational load of high-dimensional satellite data while maintaining classification accuracy for real-time processing.\"}]","Kalman-Enhanced Streaming Machine Learning for Real-Time Land Use Classification in Satellite Imagery | PDF",1785676139,272,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"kalman-enhanced-streaming-machine-learning-for-real-time-land-use-classification-in-satellite-imagery","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/kalman-enhanced-streaming-machine-learning-for-real-time-land-use-classification-in-satellite-imagery/117488/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the thesis address in satellite imagery classification?","Question",{"text":75,"@type":76},"It addresses the need for timely, real-time classification while processing continuous and unbounded incoming satellite data streams.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does Kalman-SLDA improve performance compared to traditional classifiers?",{"text":80,"@type":76},"The pipeline combines Streaming Linear Discriminant Analysis with Kalman filtering and dimensionality reduction, achieving higher accuracy and efficiency even with class imbalance and concept drift.",{"name":82,"@type":73,"acceptedAnswer":83},"Why are dimensionality reduction techniques used in the proposed pipeline?",{"text":84,"@type":76},"They reduce the computational load of high-dimensional satellite data while maintaining classification accuracy for real-time processing.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]