CNN-ViT: A multi-feature learning based approach for driver drowsiness detection
Driver drowsiness remains a critical contributor to road accidents, frequently resulting in severe injuries and fatalities. To address this issue, the present study proposes an advanced drowsiness detection system that combines the competencies of Convolutional Neural Networks (CNNs) — namely DenseN...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-09-01
|
| Series: | Array |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590005625000529 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839648355893903360 |
|---|---|
| author | Madduri Venkateswarlu Venkata Rami Reddy Chirra |
| author_facet | Madduri Venkateswarlu Venkata Rami Reddy Chirra |
| author_sort | Madduri Venkateswarlu |
| collection | DOAJ |
| description | Driver drowsiness remains a critical contributor to road accidents, frequently resulting in severe injuries and fatalities. To address this issue, the present study proposes an advanced drowsiness detection system that combines the competencies of Convolutional Neural Networks (CNNs) — namely DenseNet121, VGG16, VGG19, and ResNet50 — with a Vision Transformer (ViT). This hybrid framework is designed to harness the complementary strengths of CNNs and transformers: CNNs excel at capturing fine-grained local features, while ViT effectively models global dependencies within images. The input images are processed simultaneously through both branches, and their extracted features are merged and used to classify the driver’s state into one of four categories: Closed, Open, no_yawn, or yawn. The proposed system was evaluated on two separate datasets, named Dataset-1 and Dataset-2. Results demonstrated that the ResNet50_ViT hybrid attained a high accuracy of 99.76% on Dataset-1, while the VGG19_ViT model attained 98.21% on Dataset-2. Performance was assessed using metrics such as accuracy, precision, F1-score, and recall. The strong results, supported by optimized hyperparameters, highlight the reliability and effectiveness of the hybrid model for real-time driver drowsiness detection. |
| format | Article |
| id | doaj-art-4989bbf5f00f43e5b5cceb9da2fe7cbe |
| institution | Matheson Library |
| issn | 2590-0056 |
| language | English |
| publishDate | 2025-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Array |
| spelling | doaj-art-4989bbf5f00f43e5b5cceb9da2fe7cbe2025-06-29T04:52:47ZengElsevierArray2590-00562025-09-0127100425CNN-ViT: A multi-feature learning based approach for driver drowsiness detectionMadduri Venkateswarlu0Venkata Rami Reddy Chirra1School of Computer Science and Engineering, VIT-AP University, Amaravati, 522237, Andhra Pradesh, IndiaCorresponding author.; School of Computer Science and Engineering, VIT-AP University, Amaravati, 522237, Andhra Pradesh, IndiaDriver drowsiness remains a critical contributor to road accidents, frequently resulting in severe injuries and fatalities. To address this issue, the present study proposes an advanced drowsiness detection system that combines the competencies of Convolutional Neural Networks (CNNs) — namely DenseNet121, VGG16, VGG19, and ResNet50 — with a Vision Transformer (ViT). This hybrid framework is designed to harness the complementary strengths of CNNs and transformers: CNNs excel at capturing fine-grained local features, while ViT effectively models global dependencies within images. The input images are processed simultaneously through both branches, and their extracted features are merged and used to classify the driver’s state into one of four categories: Closed, Open, no_yawn, or yawn. The proposed system was evaluated on two separate datasets, named Dataset-1 and Dataset-2. Results demonstrated that the ResNet50_ViT hybrid attained a high accuracy of 99.76% on Dataset-1, while the VGG19_ViT model attained 98.21% on Dataset-2. Performance was assessed using metrics such as accuracy, precision, F1-score, and recall. The strong results, supported by optimized hyperparameters, highlight the reliability and effectiveness of the hybrid model for real-time driver drowsiness detection.http://www.sciencedirect.com/science/article/pii/S2590005625000529Hybrid CNN-viT modelVision transformerResNet50DenseNet121VGG19VGG16 |
| spellingShingle | Madduri Venkateswarlu Venkata Rami Reddy Chirra CNN-ViT: A multi-feature learning based approach for driver drowsiness detection Array Hybrid CNN-viT model Vision transformer ResNet50 DenseNet121 VGG19 VGG16 |
| title | CNN-ViT: A multi-feature learning based approach for driver drowsiness detection |
| title_full | CNN-ViT: A multi-feature learning based approach for driver drowsiness detection |
| title_fullStr | CNN-ViT: A multi-feature learning based approach for driver drowsiness detection |
| title_full_unstemmed | CNN-ViT: A multi-feature learning based approach for driver drowsiness detection |
| title_short | CNN-ViT: A multi-feature learning based approach for driver drowsiness detection |
| title_sort | cnn vit a multi feature learning based approach for driver drowsiness detection |
| topic | Hybrid CNN-viT model Vision transformer ResNet50 DenseNet121 VGG19 VGG16 |
| url | http://www.sciencedirect.com/science/article/pii/S2590005625000529 |
| work_keys_str_mv | AT maddurivenkateswarlu cnnvitamultifeaturelearningbasedapproachfordriverdrowsinessdetection AT venkataramireddychirra cnnvitamultifeaturelearningbasedapproachfordriverdrowsinessdetection |