{"id":{"repo_id":"mtu","oai_identifier":"oai:digitalcommons.mtu.edu:etdr-1919"},"canonical_url":"https://search.ndltd.org/etd/mtu/oai:digitalcommons.mtu.edu:etdr-1919","repository":{"repo_id":"mtu","name":"Michigan Technological University","base_url":"https://digitalcommons.mtu.edu/do/oai/"},"display":{"title":"COMPARATIVE STUDY OF HOG AND YOLO FOR TRAFFIC SIGN DETECTION SYSTEMS","abstract":"Modern automobiles have an ability to make the driving process autonomous and safer. Advanced Driver Assistance Systems (ADAS) is a cutting-edge technology that deals with the development of subsystems like traffic sign recognition to advance the driving experience and avoid human errors made while commuting. Through the collaboration of artificial intelligence with automotive industry, various algorithms have been developed to improve the performance of traffic Sign Recognition systems. However, there is an ambiguity in selecting the most efficient algorithm for designing such systems due to variations in experimental platforms and lack of common datasets. The goal of this study is to compare two algorithms – Histogram of Oriented Gradient (HOG) with linear support vector machine (SVM) and You Look Only Once (YOLO) on same hardware platform using dataset provided by Laboratory for Intelligent and Safe Automobiles (LISA) and analyze which algorithm performs effectively to identify US traffic signs. In order to compare these algorithms, performance parameters such as detection accuracy of traffic signs, misprediction rate, speed of processing the algorithm are considered. The mentioned algorithms were tested on a CPU. Postprocessing the results it was identified that, if the Traffic Sign Recognition system being developed is more concerned about processing speed then HOG with linear SVM based system can be developed. On the other hand, if the priority of detection accuracy supersedes the processing speed then YOLO based system is the ideal choice. Regardless of the performance outcomes, these algorithms are sensitive to the variations in illumination levels, an occluded region of interest, changes in shapes, size of the signs and also the viewing direction. Future scope for addressing these limitations would be employing better quality image datasets and using advanced processors like GPU, FPGA to achieve the balance between detection accuracy and speed.","abstract_html":"Modern automobiles have an ability to make the driving process autonomous and safer. Advanced Driver Assistance Systems (ADAS) is a cutting-edge technology that deals with the development of subsystems like traffic sign recognition to advance the driving experience and avoid human errors made while commuting. Through the collaboration of artificial intelligence with automotive industry, various algorithms have been developed to improve the performance of traffic Sign Recognition systems. However, there is an ambiguity in selecting the most efficient algorithm for designing such systems due to variations in experimental platforms and lack of common datasets. The goal of this study is to compare two algorithms – Histogram of Oriented Gradient (HOG) with linear support vector machine (SVM) and You Look Only Once (YOLO) on same hardware platform using dataset provided by Laboratory for Intelligent and Safe Automobiles (LISA) and analyze which algorithm performs effectively to identify US traffic signs. In order to compare these algorithms, performance parameters such as detection accuracy of traffic signs, misprediction rate, speed of processing the algorithm are considered. The mentioned algorithms were tested on a CPU. Postprocessing the results it was identified that, if the Traffic Sign Recognition system being developed is more concerned about processing speed then HOG with linear SVM based system can be developed. On the other hand, if the priority of detection accuracy supersedes the processing speed then YOLO based system is the ideal choice. Regardless of the performance outcomes, these algorithms are sensitive to the variations in illumination levels, an occluded region of interest, changes in shapes, size of the signs and also the viewing direction. Future scope for addressing these limitations would be employing better quality image datasets and using advanced processors like GPU, FPGA to achieve the balance between detection accuracy and speed.","abstract_has_math":false,"creators":["More, Sphurti"],"institution":null,"degree_name":"Master of Science in Electrical Engineering (MS)","degree_level":"Campus Access Master's Thesis","degree_discipline":"Department of Electrical and Computer Engineering","degree_department":null,"school":null,"contributors":["Jeremy P. Bos"],"advisors":[],"committee_chairs":[],"committee_members":["Michael C. Roggemann","Anthony Pinar"],"year":null,"date_issued":"","date_published":null,"updated_at":"2026-11-19T08:17:40Z","subjects":["Object detection","neural networks","traffic sign detection","ADAS","SVM","Computer Engineering","Electrical and Computer Engineering","Robotics"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.mtu.edu/etdr/818"],"render_values":[{"text":"https://digitalcommons.mtu.edu/etdr/818","href":"https://digitalcommons.mtu.edu/etdr/818","code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/10.37099/mtu.dc.etdr/818","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/10.37099/mtu.dc.etdr/818","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://digitalcommons.mtu.edu/do/oai/?verb=GetRecord&metadataPrefix=oai_etdms&identifier=oai%3Adigitalcommons.mtu.edu%3Aetdr-1919","prefix":"oai_etdms"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Jeremy P. Bos"]},{"key":"dc:contributor.committeemember","label":"Committee Member","values":["Michael C. Roggemann","Anthony Pinar"]},{"key":"dc:creator","label":"Author","values":["More, Sphurti"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.created","label":"Dc Date Created","values":["2019-01-01T08:00:00Z"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Department of Electrical and Computer Engineering"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Campus Access Master's Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science in Electrical Engineering (MS)"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Object detection","neural networks","traffic sign detection","ADAS","SVM","Computer Engineering","Electrical and Computer Engineering","Robotics"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["10.37099/mtu.dc.etdr/818","https://digitalcommons.mtu.edu/etdr/818"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["Modern automobiles have an ability to make the driving process autonomous and safer. Advanced Driver Assistance Systems (ADAS) is a cutting-edge technology that deals with the development of subsystems like traffic sign recognition to advance the driving experience and avoid human errors made while commuting. Through the collaboration of artificial intelligence with automotive industry, various algorithms have been developed to improve the performance of traffic Sign Recognition systems. However, there is an ambiguity in selecting the most efficient algorithm for designing such systems due to variations in experimental platforms and lack of common datasets. The goal of this study is to compare two algorithms – Histogram of Oriented Gradient (HOG) with linear support vector machine (SVM) and You Look Only Once (YOLO) on same hardware platform using dataset provided by Laboratory for Intelligent and Safe Automobiles (LISA) and analyze which algorithm performs effectively to identify US traffic signs. In order to compare these algorithms, performance parameters such as detection accuracy of traffic signs, misprediction rate, speed of processing the algorithm are considered. The mentioned algorithms were tested on a CPU. Postprocessing the results it was identified that, if the Traffic Sign Recognition system being developed is more concerned about processing speed then HOG with linear SVM based system can be developed. On the other hand, if the priority of detection accuracy supersedes the processing speed then YOLO based system is the ideal choice. Regardless of the performance outcomes, these algorithms are sensitive to the variations in illumination levels, an occluded region of interest, changes in shapes, size of the signs and also the viewing direction. Future scope for addressing these limitations would be employing better quality image datasets and using advanced processors like GPU, FPGA to achieve the balance between detection accuracy and speed."]},{"key":"dc:title","label":"Title","values":["COMPARATIVE STUDY OF HOG AND YOLO FOR TRAFFIC SIGN DETECTION SYSTEMS"]}]}],"canonical_facts":{"dc:contributor":["Jeremy P. Bos"],"dc:contributor.committeemember":["Michael C. Roggemann","Anthony Pinar"],"dc:creator":["More, Sphurti"],"dc:date.created":["2019-01-01T08:00:00Z"],"dc:description.abstract":["Modern automobiles have an ability to make the driving process autonomous and safer. Advanced Driver Assistance Systems (ADAS) is a cutting-edge technology that deals with the development of subsystems like traffic sign recognition to advance the driving experience and avoid human errors made while commuting. Through the collaboration of artificial intelligence with automotive industry, various algorithms have been developed to improve the performance of traffic Sign Recognition systems. However, there is an ambiguity in selecting the most efficient algorithm for designing such systems due to variations in experimental platforms and lack of common datasets. The goal of this study is to compare two algorithms – Histogram of Oriented Gradient (HOG) with linear support vector machine (SVM) and You Look Only Once (YOLO) on same hardware platform using dataset provided by Laboratory for Intelligent and Safe Automobiles (LISA) and analyze which algorithm performs effectively to identify US traffic signs. In order to compare these algorithms, performance parameters such as detection accuracy of traffic signs, misprediction rate, speed of processing the algorithm are considered. The mentioned algorithms were tested on a CPU. Postprocessing the results it was identified that, if the Traffic Sign Recognition system being developed is more concerned about processing speed then HOG with linear SVM based system can be developed. On the other hand, if the priority of detection accuracy supersedes the processing speed then YOLO based system is the ideal choice. Regardless of the performance outcomes, these algorithms are sensitive to the variations in illumination levels, an occluded region of interest, changes in shapes, size of the signs and also the viewing direction. Future scope for addressing these limitations would be employing better quality image datasets and using advanced processors like GPU, FPGA to achieve the balance between detection accuracy and speed."],"dc:identifier":["10.37099/mtu.dc.etdr/818","https://digitalcommons.mtu.edu/etdr/818"],"dc:subject":["Object detection","neural networks","traffic sign detection","ADAS","SVM","Computer Engineering","Electrical and Computer Engineering","Robotics"],"dc:title":["COMPARATIVE STUDY OF HOG AND YOLO FOR TRAFFIC SIGN DETECTION SYSTEMS"],"thesis:degree_discipline":["Department of Electrical and Computer Engineering"],"thesis:degree_level":["Campus Access Master's Thesis"],"thesis:degree_name":["Master of Science in Electrical Engineering (MS)"]},"updated_at":"2026-11-19T08:17:40Z"}