[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120155-en":3,"doc-seo-120155-105":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},120155,8796095461610,"Oliver","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Automated Classification of Irregularities in Magnetic Audio Tapes Using Various Machine Learning Techniques","This thesis presents an automated approach to classifying irregularities on magnetic audio tapes using deep learning. It compares multiple image preprocessing strategies, including thresholded images, difference images, opened images, and Region of Interest (ROI) images, to determine which representation best supports recognition. A ResNet50 model is used as the core classifier, with extensive experiments and fine-tuning. Results show ROI images achieve the highest test accuracy (about 98.37%) by emphasizing informative regions and reducing irrelevant background effects. Data imbalance and overfitting are mitigated via augmentation and careful tuning, and future work proposes GAN-based synthetic data, enhanced augmentation, and ensemble learning.","DIPARTIMENTO DI INGEGNERIA DELL’INFORMAZIONE  \nCORSO DI LAUREA MAGISTRALE IN ICT FOR INTERNET AND MULTIMEDIA  \n“Automated Classification of  \nIrregularities in Magnetic Audio Tapes  \nUsing Various Machine Learning  \nTechniques  \n”  \nRelatore: Prof. / Dott Sergia Canazza Targon  \nLaureando/a: Mehmet Ozturk  \nCorrelatore: Prof,/Dott Matteo Spanio , Alessandro Russo  \nANNO ACCADEMICO 2023.– 2024..  \nData di laurea 15/10/2024  \nUniversity of Padova  \nDepartment of Information Engineering Master Thesis in ICT for Internet and Multimedia  \nAutomated Classification of Irregularities in Magnetic Audio Tapes Using Various Machine Learning  \nTechniques  \nSupervisor Master Candidate  \nProf. Sergio Canazza Targon Mehmet Ozturk  \nUniversity of Padova  \nCo-supervisor Student ID  \nMatteo Spanio 2049527  \nAlessandro Russo  \nAcademic Year  \n2021-2024  \nii  \nTo my fiancée, Iremsu Savas, whose love, support, and understanding have been my guiding light throughout this journey. This thesis is as much yours as it is mine.  \nI would like to express my sincere gratitude tomy professor, Sergio Canazza Targon, for his invaluable guidance and expertise throughout this project. My heartfelt thanks also go to my supervisors, Alessandro Russo and Matteo Spanio, for their unwavering support and insightful feedback. Iam deeply  \nappreciative of Zafer Cinar for his collaboration and constant encouragement. Your contributions have been crucial to the completion of this thesis.  \niv  \nAbstract  \nThis thesis presents a comprehensive approach to the classification of magnetic tape irregularities using deep learning techniques, focusing on various image preprocessing methods to enhance model performance. The primary objective was to evaluate the effectiveness of different datasets, including thresholded images, difference images, opened images, and Region of Interest (ROI) images, in identifying specific types of irregularities. The ResNet50 architecture was employed as the core model due to its proven efficacy in image classification tasks.  \nThrough extensive experimentation, the ROI images dataset emerged as the most effective approach, yielding the highest test accuracy of approximately 98.37% . This superior performance can be attributed to the model’s ability to focus on the most pertinent regions of the images, thereby enhancing the feature extraction process and minimizing the influence of irrelevant background noise. In contrast, other datasets such as thresholded and difference images showed a decline in performance, highlighting the importance of careful preprocessing in achieving accurate classification.  \nChallenges such as data imbalance and model overfitting were addressed through techniques like data augmentation and careful model fine-tuning. However, some limitations persisted, including the need for further optimization and the exploration of additional data representations. The study proposes future work directions, including the use of Generative Adversarial Networks (GANs) for synthetic data generation, advanced augmentation methods, and ensemble learning to further improve classification accuracy and robustness.  \nIn conclusion, this research demonstrates the critical role of targeted image preprocessing in enhancing deep learning models for specific classification tasks. The insights gained from this study provide a foundation for future advancements in the automated detection and classification of magnetic tape irregularities, with potential applications in media preservation and quality control.  \nvi  \nContents  \nAbstract v  \nList of figures xi  \nList of tables xiii  \n1 Introduction 3  \n1.1 Image Classification .............................. 4  \n1.2 Challenges in Image Classification and Their Solutions ............ 5  \n1.3 Audio Preservation and Digitization ..................... 6  \n1.4 Centro di Sonologia Computazionale ..................... 7  \n1.5 MPAI-CAE ARP ............................... 8  \n1.5.1 Moving Picture, Audio and Data ","cbCaigyplP9x17Sg","https://ap.wps.com/l/cbCaigyplP9x17Sg","pdf",3215832,1,109,"English","en",105,"# Introduction\n## Image Classification\n## Challenges in Image Classification and Their Solutions\n## Audio Preservation and Digitization\n## Classification of Irregularities\n# Tape Irregularity Classification Dataset\n## Acquisition of Dataset (Video Analyser)\n## Dataset Content\n## Data Augmentation Methods\n# Models\n## Convolutional Neural Network (CNN) Classification Approach\n## Selecting the Best Model\n## yolo v7","[{\"question\":\"Which image preprocessing method performs best for irregularity classification?\",\"answer\":\"The ROI images dataset yields the highest test accuracy of approximately 98.37%, outperforming thresholded and difference image approaches.\"},{\"question\":\"What model architecture is used in the thesis?\",\"answer\":\"ResNet50 is employed as the core model because it has proven effective for image classification tasks.\"},{\"question\":\"How are challenges like data imbalance and overfitting handled?\",\"answer\":\"The thesis addresses them using data augmentation and careful model fine-tuning, and it also highlights remaining limitations that motivate further optimization.\"}]","Automated Classification of Irregularities in Magnetic Audio Tapes Using Various Machine Learning Techniques | PDF",1785728469,275,{"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},"automated-classification-of-irregularities-in-magnetic-audio-tapes-using-various-machine-learning-techniques","",{"@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/automated-classification-of-irregularities-in-magnetic-audio-tapes-using-various-machine-learning-techniques/120155/",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","2026-08-04","2026-08-03",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},"Which image preprocessing method performs best for irregularity classification?","Question",{"text":76,"@type":77},"The ROI images dataset yields the highest test accuracy of approximately 98.37%, outperforming thresholded and difference image approaches.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What model architecture is used in the thesis?",{"text":81,"@type":77},"ResNet50 is employed as the core model because it has proven effective for image classification tasks.",{"name":83,"@type":74,"acceptedAnswer":84},"How are challenges like data imbalance and overfitting handled?",{"text":85,"@type":77},"The thesis addresses them using data augmentation and careful model fine-tuning, and it also highlights remaining limitations that motivate further optimization.","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"]