[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122130-en":3,"doc-seo-122130-105":30,"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":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},122130,1099514068035,"Ezra","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Hardware implementation of machine learning algorithms for image processing - Master’s thesis","Image processing drives innovation across medicine, security, and industrial automation, with classification forming a key basis for identifying and categorizing objects and visual patterns. This master’s thesis investigates a hardware implementation of machine learning image-classification algorithms. After reviewing machine learning methods for image processing, the Deep Forest algorithm was selected for high accuracy with low computational load. A Python model served as a golden reference to optimize parameters, then Vitis High Level Synthesis mapped the algorithm to FPGA hardware, achieving up to 50x CPU speedup and about 98% accuracy on single-band and hyperspectral images.","Master’s thesis  \nLuis Valls Sansaloni  \nHardware implementation of machine learning algorithms for image processing  \nMaster’s thesis in European Masters programme in Embedded Computing Systems (EMECS)  \nSupervisor: Milica Orlandic  \nCo-supervisor: Samuel Boyle and Dordije Boskovic June 2024  \nNT NU  \nNorwegian Un iversity of Science and Technology  \nFaculty of Information Techno logy and Electrical Engineering Department of Electronic Systems  \nLuis Valls Sansaloni  \nHardware implementation of machine learning algorithms for image processing  \nMaster’s thesis in European Masters programme in Embedded Computing Systems (EMECS)  \nSupervisor: Milica Orlandic  \nCo-supervisor: Samuel Boyle and Dordije Boskovic June 2024  \nNorwegian University of Science and Technology  \nFaculty of Information Technology and Electrical Engineering Department of Electronic Systems  \nABSTRACT  \nImage processing is essential nowadays, driving advancements in fields like medicine, security, and industrial automation. Among the various algorithms used in image processing, classification serves as the foundation for many other systems, enabling the automatic identification and categorization of objects and visual patterns.  \nThis master’s thesis explores the hardware implementation of image processing algorithms that utilize machine learning, specifically targeting image classification. After an initial research into machine learning algorithms applied to image processing, the Deep Forest algorithm was chosen for its potential to achieve high accuracy with a low computational load. A Python model was created to serve asa golden reference and to optimize the algorithm’s parameters. Vitis High Level Synthesis was then used to translate the algorithm into a hardware model. Three different hardware architectures were implemented on an Xilinx AMD FPGA device, resulting in a performance improvement of factor 50 times faster compared to CPU implementations. The Deep Forest algorithm demonstrated high accuracy in processing both single-band and hyperspectral images, achieving accuracy levels of approximately 98% .  \nBildeprosessering er en essensiell teknologi i dagens samfunn, og driver innovasjon innenfor bransjer som medisin, sikkerhet, og industriell automasjon. Klassifisering, en av mange algoritmer brukt i bildeprosessering, er det mange bildeanalysesystemer er bygget på, og muliggjør automatisk identifisering og kategorisering av objekter og mønstre.  \nDenne masteroppgaven utforsker hardvareimplementasjonen av bildeprosesseringsalgoritmer som bruker maskinlæring innenfor bildeklassifisering. Etter et litteratursøk innenfor maskinlæringsalgoritmer for bildeprosessering ble Deep Forestalgoritmen valgt på grunn av dens potensiale for høy presisjon med en lav mengde operasjoner. En Python-modell ble laget for å brukes som en perfekt referanse og for å optimalisere algoritmens parametre. Vitis High Level Synthesis ble så brukt for å oversette algoritmen til en hardvaremodell. Tre ulike arkitekturer ble implementert på en AMD-Xilinx FPGA, som resulterte i en 50 ganger høyere hastighetenn CPU-baserte implementasjoner. Deep Forest-algoritmen viste høy presisjon iprosessering av både enkeltbånds-og hyperspektrale bilder, med et presisjonsnivå på omtrent 98% .  \nACKNOWLEDGEMENTS  \nI want to start by thanking my family. Your constant support and encouragement have been the foundation of my thesis. Balancing a part-time job and working on this project was tough, but your belief in me kept me going. Thank you for always being there, offering advice, and for your endless patience and love. Without you, I couldn’t have done this.  \nI also want to thank my friends all around the globe, especially in Trondheim. You helped me de-stress and have fun when I needed a break from work. The time we spent together was awesome and gave me the energy to continue. I’m so grateful for your friendship and all the amazing memories we made.  \nMilica, a special thanks to you for giving ","cbCaijc91JuTD7yU","https://ap.wps.com/l/cbCaijc91JuTD7yU","pdf",15626667,1,67,"English","en",105,"# Abstract\n# Acknowledgements\n# Contents\n# List of Figures\n# List of Tables\n# Abbreviations\n# Introduction\n## Motivation\n## Objectives\n## Thesis Organization\n# Background\n## Images and bands\n## Image processing\n## Artificial Intelligence, Machine Learning and Deep Learning\n## Machine learning algorithms applied to image processing\n## High Language Synthesis\n## Conclusions\n# Deep Forest algorithm and Python model\n## Deep Forest algorithm description\n## Python implementation\n## Verification of the Python models\n# Hardware implementation\n## The C implementation","[{\"question\":\"Why is image classification important in image processing systems?\",\"answer\":\"Classification is a foundational component that enables automatic identification and categorization of objects and visual patterns.\"},{\"question\":\"How was the Deep Forest algorithm selected for this thesis?\",\"answer\":\"It was chosen after research into relevant machine learning algorithms, mainly for its potential to achieve high accuracy with low computational load.\"},{\"question\":\"What hardware/software tools were used to implement the algorithm on FPGA?\",\"answer\":\"A Python golden-reference model was built first, and Vitis High Level Synthesis was used to translate the algorithm into FPGA hardware implementations.\"}]","Hardware implementation of machine learning algorithms for image processing - Master’s thesis | PDF",1785808966,169,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"hardware-implementation-of-machine-learning-algorithms-for-image-processing-masters-thesis","",{"@graph":36,"@context":86},[37,54,69],{"@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/hardware-implementation-of-machine-learning-algorithms-for-image-processing-masters-thesis/122130/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-04",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 image classification important in image processing systems?","Question",{"text":76,"@type":77},"Classification is a foundational component that enables automatic identification and categorization of objects and visual patterns.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How was the Deep Forest algorithm selected for this thesis?",{"text":81,"@type":77},"It was chosen after research into relevant machine learning algorithms, mainly for its potential to achieve high accuracy with low computational load.",{"name":83,"@type":74,"acceptedAnswer":84},"What hardware/software tools were used to implement the algorithm on FPGA?",{"text":85,"@type":77},"A Python golden-reference model was built first, and Vitis High Level Synthesis was used to translate the algorithm into FPGA hardware implementations.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]