{"id":{"repo_id":"vt","oai_identifier":"oai:vtechworks.lib.vt.edu:10919/123787"},"canonical_url":"https://search.dev.ndltd.org/etd/vt/oai:vtechworks.lib.vt.edu:10919/123787","repository":{"repo_id":"vt","name":"Virginia Tech","base_url":"https://vtechworks.lib.vt.edu/oai/request"},"display":{"title":"AI-Driven Pig Monitoring System: Behavior and Weight Analysis","abstract":"This thesis advances automated pig monitoring through novel machine learning approaches in behavior analysis, weight prediction and forecasting. For behavior analysis, we introduce a preprocessing framework that addresses data leakage in time series analysis through non-class-based windowing and chronological sampling, achieving up to 15% improvement in accuracy over conventional methods. For current weight prediction, we develop an automated pipeline using the Segment Anything Model (SAM) with deep learning, where our Xception-Net architecture achieves a mean absolute percentage error of 7.42%. For weight forecasting, we propose multi-input deep learning architectures combining spatial and temporal features, achieving a mean absolute percentage error of 5.56%. These methods demonstrate robust performance in real-world conditions while minimizing animal stress and manual labor requirements, contributing significantly to precision livestock farming practices.","abstract_html":"This thesis advances automated pig monitoring through novel machine learning approaches in behavior analysis, weight prediction and forecasting. For behavior analysis, we introduce a preprocessing framework that addresses data leakage in time series analysis through non-class-based windowing and chronological sampling, achieving up to 15% improvement in accuracy over conventional methods. For current weight prediction, we develop an automated pipeline using the Segment Anything Model (SAM) with deep learning, where our Xception-Net architecture achieves a mean absolute percentage error of 7.42%. For weight forecasting, we propose multi-input deep learning architectures combining spatial and temporal features, achieving a mean absolute percentage error of 5.56%. These methods demonstrate robust performance in real-world conditions while minimizing animal stress and manual labor requirements, contributing significantly to precision livestock farming practices.","abstract_has_math":false,"creators":["Ranjan, Pranjal"],"institution":"Virginia Tech","degree_name":"Master of Science","degree_level":"masters","degree_discipline":"Computer Engineering","degree_department":"Electrical and Computer Engineering","school":null,"contributors":[],"advisors":[],"committee_chairs":["Shin, Sook"],"committee_members":["Morota, Gota","Ha, Dong S."],"year":2024,"date_issued":"2024-12-12","date_published":"2024-12-12","updated_at":"2026-07-22T22:19:53Z","subjects":["Artificial Intelligence","Deep Learning","Precision Livestock Farming"],"languages":["en"],"rights":["In Copyright"],"rights_urls":["http://rightsstatements.org/vocab/InC/1.0/"],"identifier_entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:42077"],"render_values":[{"text":"vt_gsexam:42077","href":null,"code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/10919/123787","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.committeechair","label":"Committee Chair","values":["Shin, Sook"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Morota, Gota","Ha, Dong S."]},{"key":"dc:contributor.department","label":"Department","values":["Electrical and Computer Engineering"]},{"key":"dc:creator","label":"Author","values":["Ranjan, Pranjal"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2024-12-13T09:00:11Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2024-12-13T09:00:11Z"]},{"key":"dc:date.issued","label":"Date","values":["2024-12-12"]},{"key":"dc:publisher","label":"Institution","values":["Virginia Tech"]},{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["masters"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["Virginia Polytechnic Institute and State University"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Artificial Intelligence","Deep Learning","Precision Livestock Farming"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["In Copyright"]},{"key":"dc:rights.uri","label":"Rights URI","values":["http://rightsstatements.org/vocab/InC/1.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.other","label":"Dc Identifier Other","values":["vt_gsexam:42077"]},{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/10919/123787"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis advances automated pig monitoring through novel machine learning approaches in behavior analysis, weight prediction and forecasting. For behavior analysis, we introduce a preprocessing framework that addresses data leakage in time series analysis through non-class-based windowing and chronological sampling, achieving up to 15% improvement in accuracy over conventional methods. For current weight prediction, we develop an automated pipeline using the Segment Anything Model (SAM) with deep learning, where our Xception-Net architecture achieves a mean absolute percentage error of 7.42%. For weight forecasting, we propose multi-input deep learning architectures combining spatial and temporal features, achieving a mean absolute percentage error of 5.56%. These methods demonstrate robust performance in real-world conditions while minimizing animal stress and manual labor requirements, contributing significantly to precision livestock farming practices."]},{"key":"dc:description.abstractgeneral","label":"General Abstract","values":["Modern pig farming faces increasing pressure to efficiently monitor animal health and growth while ensuring high welfare standards. This research develops smart computer systems that can automatically track three important aspects of pig farming: how pigs behave, how much they currently weigh, and how much they will weigh in the future. Instead of requiring farmers to physically handle pigs for weighing or spending hours observing their behavior, our system uses cameras and sensors to collect this information automatically. We create new computer programs that can recognize different pig behaviors like eating, sleeping, and walking with over 95% accuracy. For weight monitoring, we develop a system using special depth-sensing cameras that can estimate a pig's weight within 7% of their actual weight, all without needing to move or disturb the animal. Looking ahead, our system can also predict future pig weights with over 94% accuracy, helping farmers make better decisions about feeding and care. These tools significantly reduce the time and effort needed for monitoring pigs while decreasing animal stress from handling. By providing accurate, real-time information about pig behavior and growth, this research helps farmers make better management decisions, ultimately leading to improved animal welfare and more efficient farming operations."]},{"key":"dc:description.degree","label":"Dc Description Degree","values":["Master of Science"]},{"key":"dc:format.medium","label":"Dc Format Medium","values":["ETD"]},{"key":"dc:title","label":"Title","values":["AI-Driven Pig Monitoring System: Behavior and Weight Analysis"]}]}],"canonical_facts":{"dc:contributor.committeechair":["Shin, Sook"],"dc:contributor.committeemember":["Morota, Gota","Ha, Dong S."],"dc:contributor.department":["Electrical and Computer Engineering"],"dc:creator":["Ranjan, Pranjal"],"dc:date.accessioned":["2024-12-13T09:00:11Z"],"dc:date.available":["2024-12-13T09:00:11Z"],"dc:date.issued":["2024-12-12"],"dc:description.abstract":["This thesis advances automated pig monitoring through novel machine learning approaches in behavior analysis, weight prediction and forecasting. For behavior analysis, we introduce a preprocessing framework that addresses data leakage in time series analysis through non-class-based windowing and chronological sampling, achieving up to 15% improvement in accuracy over conventional methods. For current weight prediction, we develop an automated pipeline using the Segment Anything Model (SAM) with deep learning, where our Xception-Net architecture achieves a mean absolute percentage error of 7.42%. For weight forecasting, we propose multi-input deep learning architectures combining spatial and temporal features, achieving a mean absolute percentage error of 5.56%. These methods demonstrate robust performance in real-world conditions while minimizing animal stress and manual labor requirements, contributing significantly to precision livestock farming practices."],"dc:description.abstractgeneral":["Modern pig farming faces increasing pressure to efficiently monitor animal health and growth while ensuring high welfare standards. This research develops smart computer systems that can automatically track three important aspects of pig farming: how pigs behave, how much they currently weigh, and how much they will weigh in the future. Instead of requiring farmers to physically handle pigs for weighing or spending hours observing their behavior, our system uses cameras and sensors to collect this information automatically. We create new computer programs that can recognize different pig behaviors like eating, sleeping, and walking with over 95% accuracy. For weight monitoring, we develop a system using special depth-sensing cameras that can estimate a pig's weight within 7% of their actual weight, all without needing to move or disturb the animal. Looking ahead, our system can also predict future pig weights with over 94% accuracy, helping farmers make better decisions about feeding and care. These tools significantly reduce the time and effort needed for monitoring pigs while decreasing animal stress from handling. By providing accurate, real-time information about pig behavior and growth, this research helps farmers make better management decisions, ultimately leading to improved animal welfare and more efficient farming operations."],"dc:description.degree":["Master of Science"],"dc:format.medium":["ETD"],"dc:identifier.other":["vt_gsexam:42077"],"dc:identifier.uri":["https://hdl.handle.net/10919/123787"],"dc:language.iso":["en"],"dc:publisher":["Virginia Tech"],"dc:rights":["In Copyright"],"dc:rights.uri":["http://rightsstatements.org/vocab/InC/1.0/"],"dc:subject":["Artificial Intelligence","Deep Learning","Precision Livestock Farming"],"dc:title":["AI-Driven Pig Monitoring System: Behavior and Weight Analysis"],"dc:type":["Thesis"],"thesis:degree_discipline":["Computer Engineering"],"thesis:degree_level":["masters"],"thesis:degree_name":["Master of Science"],"thesis:institution_name":["Virginia Polytechnic Institute and State University"]},"updated_at":"2026-07-22T22:19:53Z"}