Distracted driver detection with deep convolution neural networks
This project aims to develop a python algorithm to detect the distracted activities while driving. National Highway traffic Safety Administration of United State (NHTSA) has been reported in 2015 around 3477 deaths cases and injuries to 391000 people because of distracted driving, with that distract...
Saved in:
| Main Author: | |
|---|---|
| Format: | text::Final Year Project |
| Language: | en_US |
| Published: |
2023
|
| Subjects: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1840843679017730048 |
|---|---|
| author | Basubeit, Omar Gumaan Saleh |
| author_facet | Basubeit, Omar Gumaan Saleh |
| author_sort | Basubeit, Omar Gumaan Saleh |
| building | UNITEN Library |
| collection | Institutional Repository |
| content_provider | Universiti Tenaga Nasional |
| content_source | UNITEN Institutional Repository |
| continent | Asia |
| country | Malaysia |
| description | This project aims to develop a python algorithm to detect the distracted activities while driving. National Highway traffic Safety Administration of United State (NHTSA) has been reported in 2015 around 3477 deaths cases and injuries to 391000 people because of distracted driving, with that distracted driving considered as one of the main causes of car accidents. This project focuses to reduce manual and visual distractions by using visual dataset of 20,000 images divided to three sets: train set, validation set and test set. Unsafe activities are texting and talking on mobile phone, operating the radio, reaching behind to grab something, talking to passenger, drinking or eating and hair or makeup. To ensure having a high accuracy, the author decided to use deep learning technology and convolution neural networks (CNNs) in specific. The author created his algorithm and 12 Keras pre-trained models such as VGG16 and Xception, with adding 5 layers on the top of them. Moreover, this project compares Keras pre-trained models for two strategies: train only the classifier and train the top layers including the classifier. |
| format | Resource Types::text::Final Year Project |
| id | my.uniten.dspace-27654 |
| institution | Universiti Tenaga Nasional |
| language | en_US |
| publishDate | 2023 |
| record_format | dspace |
| spelling | my.uniten.dspace-276542025-08-13T09:26:26Z Distracted driver detection with deep convolution neural networks Basubeit, Omar Gumaan Saleh Neural networks (Computer science) Detectors Distracted driving This project aims to develop a python algorithm to detect the distracted activities while driving. National Highway traffic Safety Administration of United State (NHTSA) has been reported in 2015 around 3477 deaths cases and injuries to 391000 people because of distracted driving, with that distracted driving considered as one of the main causes of car accidents. This project focuses to reduce manual and visual distractions by using visual dataset of 20,000 images divided to three sets: train set, validation set and test set. Unsafe activities are texting and talking on mobile phone, operating the radio, reaching behind to grab something, talking to passenger, drinking or eating and hair or makeup. To ensure having a high accuracy, the author decided to use deep learning technology and convolution neural networks (CNNs) in specific. The author created his algorithm and 12 Keras pre-trained models such as VGG16 and Xception, with adding 5 layers on the top of them. Moreover, this project compares Keras pre-trained models for two strategies: train only the classifier and train the top layers including the classifier. 2023-07-21T01:16:41Z 2023-07-21T01:16:41Z 2019 Resource Types::text::Final Year Project https://irepository.uniten.edu.my/handle/123456789/27654 en_US application/pdf |
| spellingShingle | Neural networks (Computer science) Detectors Distracted driving Basubeit, Omar Gumaan Saleh Distracted driver detection with deep convolution neural networks |
| title | Distracted driver detection with deep convolution neural networks |
| title_full | Distracted driver detection with deep convolution neural networks |
| title_fullStr | Distracted driver detection with deep convolution neural networks |
| title_full_unstemmed | Distracted driver detection with deep convolution neural networks |
| title_short | Distracted driver detection with deep convolution neural networks |
| title_sort | distracted driver detection with deep convolution neural networks |
| topic | Neural networks (Computer science) Detectors Distracted driving |
| url_provider | http://dspace.uniten.edu.my/ |