[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120901-en":3,"doc-seo-120901-105":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":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},120901,4810365810221,"Aurora","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Enhancing Driver Distraction Detection Through the Synergy of Deep and Traditional Machine Learning - Master’s Project","Distracted driving significantly contributes to motor vehicle accidents, influencing overall driving behavior and causing injury and loss of life. This work proposes an ML-driven approach to reduce insurance-related bias by analyzing driving behavior and estimating accident risk. Using transfer learning with ResNet50 and VGG16, it performs multiclass driver distraction classification with reported accuracies up to 94.1% and 92.2%, respectively. A custom CNN reaches 98.7% validation accuracy. To improve practicality, the task is reformulated as binary classification by fusing deep feature extraction with Random Forest, achieving 99.6% validation accuracy and computing a driver-specific distraction score for safer driving and more equitable premiums.","San Jose State University  \nSJSU ScholarWorks  \n\n| Master's Projects | Theses and Graduate Research |\n| --- | --- |\n| Fall 2023\u003Cbr>Enhancing Driver Distraction Detection Through the Synergy of Deep and Traditional Machine Learning\u003Cbr>Gowtham Chandrasekaran\u003Cbr>Follow this and additional works at: [https://scholarworks.sjsu.edu/etd_projects](https://scholarworks.sjsu.edu/etd_projects)\u003Cbr> Part of the Other Computer Engineering Commons |  |\n\nRecommended Citation  \nChandrasekaran, Gowtham, \"Enhancing Driver Distraction Detection Through the Synergy of Deep and Traditional Machine Learning\" (2023) . Master 's Projects. 1308.  \nDOI: [https://doi.org/10.31979/etd.v88q-qmjn](https://doi.org/10.31979/etd.v88q-qmjn)  \n[https://scholarworks.sjsu.edu/etd_projects/1308](https://scholarworks.sjsu.edu/etd_projects/1308)  \nThis Master's Project is brought to you for free and open access by the Theses and Graduate Research at SJSU ScholarWorks. It has been accepted for inclusion in Master's Projects by an authorized administrator of SJSU ScholarWorks. For more information, please contact [scholarworks@sjsu.edu](scholarworks@sjsu.edu).  \nEnhancing Driver Distraction Detection Through the Synergy of Deep and  \nTraditional Machine Learning  \nA Project  \nPresented to  \nThe Faculty of the Department of Computer Science San José State University  \nIn Partial Fulfillment  \nof the Requirements for the Degree  \nMaster of Science  \nby  \nGowtham Chandrasekaran  \nDecember 2023  \n© 2023  \nGowtham Chandrasekaran ALL RIGHTS RESERVED  \nThe Designated Project Committee Approves the Project Titled  \nEnhancing Driver Distraction Detection Through the Synergy of Deep and  \nTraditional Machine Learning  \nby  \nGowtham Chandrasekaran  \nAPPROVED FOR THE DEPARTMENT OF COMPUTER SCIENCE  \nSAN JOSÉ STATE UNIVERSITY  \nDecember 2023  \nDr. Navrati Saxena  \nDr. Abhishek Roy  \nDr. Nada Attar  \nDepartment of Computer Science MediaTek USA Inc  \nDepartment of Computer Science  \nABSTRACT  \nEnhancing Driver Distraction Detection Through the Synergy of Deep and  \nTraditional Machine Learning  \nby Gowtham Chandrasekaran  \nDistracted driving is a major contributor to motor vehicle accidents, causing injury and loss of life. It is one of the major factors that affect the overall driving behavior of a person. Insurance companies take into consideration factors like gender, age, etc. to set insurance premiums for their customers. Today, machine learning and artificial intelligence can eradicate this bias. A machine learning model can analyze driving behavior, such as the frequency and severity of accidents, the speed at which they drive, and their habits such as distracted driving. Based on this information, the model can then determine the risk of accidents for the driver and set a corresponding insurance premium. In this project, we initiated multiclass driver distraction classification through transfer learning, utilizing pre-trained models such as ResNet50 and VGG16 to achieve 92 .2% and 94 . 1% accuracies. Subsequently, we introduced a custom CNN model which attained a validation accuracy of 98 .7% . To enhance practicality and accuracy, we transformed the multiclass task into binary classification. Here, we leveraged the fusion of deep learning and traditional machine learning, treating the convolutional filters as a feature extractor and employing Random Forest for binary classification, resulting in an impressive 99 .6% validation accuracy. Our model was trained on the State Farm Distraction dataset that is available to the public on Kaggle. Additionally, we computed a driver-specific distraction score, offering potential applications in assessing accident risk, setting equitable insurance premiums, and promoting safer driving habits.  \nKeywords: Convolutional neural network, convolutional filters, computer vision, ResNet50, VGG16, distraction score, distraction detection, distraction score  \nACKNOWLEDGMENTS  \nI’d like to thank my family first and foremost for always supporting ","cbCaiaPgsPUz2gnu","https://ap.wps.com/l/cbCaiaPgsPUz2gnu","pdf",22204385,1,82,"English","en",105,"# Chapter 1 Introduction\n## 1.1 Background\n## 1.2 Computer Vision\n## 1.3 Project Overview\n## 1.4 Report Overview\n# Chapter 2 Related Works\n## 2.1 Non CNN Approaches\n## 2.2 CNN Approaches\n# Chapter 3 Dataset\n## 3.1 Types of Distraction\n# Chapter 4 Exploratory Data Analysis\n## 4.1 Distribution of Images\n## 4.2 Number of Images per Driver\n# Chapter 5 Image Preprocessing\n## 5.1 Normalization\n## 5.2 Data Augmentation\n# Chapter 6 Transfer Learning\n# Chapter 7 Multiclass Classification\n## 7.1 Transfer Learning Using ResNet50 and VGG16\n## 7.2 Custom CNN for Multiclass Classification\n## 7.3 Custom CNN for Feature Extraction\n## 7.4 Custom CNN Results","[{\"question\":\"How does the project detect driver distraction?\",\"answer\":\"It uses machine learning to classify driver distraction from visual driving data. The approach includes transfer learning with ResNet50 and VGG16 and a custom CNN, then converts the task to binary classification for practical deployment.\"},{\"question\":\"What accuracy results are reported for the different models?\",\"answer\":\"Multiclass classification using transfer learning reports accuracies around 92.2% and 94.1%, depending on the pretrained model. The custom CNN reaches 98.7% validation accuracy, while the deep-feature + Random Forest binary fusion reports 99.6% validation accuracy.\"},{\"question\":\"What is the driver-specific distraction score used for?\",\"answer\":\"The project computes a driver-specific distraction score intended to support assessing accident risk, enabling fairer insurance premium setting, and encouraging safer driving habits.\"}]","Enhancing Driver Distraction Detection Through the Synergy of Deep and Traditional Machine Learning - Master’s Project | PDF",1785732569,207,{"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},"enhancing-driver-distraction-detection-through-the-synergy-of-deep-and-traditional-machine-learning-masters-project","",{"@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/enhancing-driver-distraction-detection-through-the-synergy-of-deep-and-traditional-machine-learning-masters-project/120901/",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","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How does the project detect driver distraction?","Question",{"text":75,"@type":76},"It uses machine learning to classify driver distraction from visual driving data. The approach includes transfer learning with ResNet50 and VGG16 and a custom CNN, then converts the task to binary classification for practical deployment.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What accuracy results are reported for the different models?",{"text":80,"@type":76},"Multiclass classification using transfer learning reports accuracies around 92.2% and 94.1%, depending on the pretrained model. The custom CNN reaches 98.7% validation accuracy, while the deep-feature + Random Forest binary fusion reports 99.6% validation accuracy.",{"name":82,"@type":73,"acceptedAnswer":83},"What is the driver-specific distraction score used for?",{"text":84,"@type":76},"The project computes a driver-specific distraction score intended to support assessing accident risk, enabling fairer insurance premium setting, and encouraging safer driving habits.","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"]