Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Data augmentation for machine learning assisted pesticide detection from SERS.

Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy·2025
Same author

Comprehensive Investigation of Machine Learning and Deep Learning Networks for Identifying Multispecies Tomato Insect Images.

Sensors (Basel, Switzerland)·2024
Same author

Anxiety in aquatics: Leveraging machine learning models to predict adult zebrafish behavior.

Neuroscience·2024
Same author

Corrigendum to 'Deep learning dives: Predicting anxiety in Zebrafish through novel tank assay analysis' Physiology & Behavior (2024), 114696.

Physiology & behavior·2024
Same author

Deep learning dives: Predicting anxiety in zebrafish through novel tank assay analysis.

Physiology & behavior·2024
Same author

Dispersive Modeling of Normal and Cancerous Cervical Cell Responses to Nanosecond Electric Fields in Reversible Electroporation Using a Drift-Step Rectifier Diode Generator.

Micromachines·2023

Related Experiment Video

Updated: Jun 18, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

512

CBAM VGG16: An efficient driver distraction classification using CBAM embedded VGG16 architecture.

Chittathuru Himala Praharsha1, Alwin Poulose1

  • 1School of Data Science, Indian Institute of Science Education and Research Thiruvananthapuram (IISER TVM), Vithura, Thiruvananthapuram 695551, Kerala, India.

Computers in Biology and Medicine
|August 2, 2024
PubMed
Summary
This summary is machine-generated.

A new CBAM VGG16 deep learning model significantly improves driver distraction classification for autonomous driving systems. This enhanced model boosts safety by accurately identifying driver activities from camera data.

Keywords:
Attention moduleAutonomous driving system (ADS)Autonomous vehiclesConvolutional neural network (CNN)Deep learningDriver distraction classificationDriver monitoring systemsImage classificationsVisual Geometry Group (VGG)

More Related Videos

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.7K
P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
06:09

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation

Published on: September 8, 2023

553

Related Experiment Videos

Last Updated: Jun 18, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

512
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

2.7K
P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
06:09

P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation

Published on: September 8, 2023

553

Area of Science:

  • Computer Vision and Machine Learning
  • Autonomous Driving Systems
  • Artificial Intelligence for Automotive Safety

Background:

  • Driver monitoring systems (DMS) are essential for autonomous driving systems (ADS) to ensure driver and vehicle safety.
  • Classifying driver distractions is critical for real-time safety enhancements in ADS.
  • Accurate driver distraction classification is challenging due to the unpredictable nature of human driving behavior.

Purpose of the Study:

  • To propose a novel deep learning architecture, CBAM VGG16, for improved driver distraction classification.
  • To enhance the feature extraction capabilities of the VGG16 model by integrating a Convolutional Block Attention Module (CBAM).
  • To evaluate the performance of the proposed CBAM VGG16 model against existing state-of-the-art architectures.

Main Methods:

  • Developed a hybrid deep learning model by embedding CBAM layers within the VGG16 architecture.
  • Trained and tested the CBAM VGG16 model on the American University in Cairo (AUC) distracted driver dataset version 2 (AUCD2).
  • Compared classification performance metrics (accuracy, loss, precision, F1 score, recall, confusion matrix) with other models like DenseNet121, Xception, MoblieNetV2, InceptionV3, and VGG16.

Main Results:

  • The proposed CBAM VGG16 achieved high classification accuracies: 98.65% for camera 1 and 97.85% for camera 2 on the AUCD2 dataset.
  • Demonstrated significant performance improvements over the conventional VGG16 model, with 3.7% and 5% increases for camera 1 and camera 2, respectively.
  • Grad-CAM visualizations confirmed that CBAM layers improve attention to critical regions in distraction images compared to VGG16 alone.

Conclusions:

  • The CBAM VGG16 architecture effectively enhances feature extraction and classification performance for driver distractions.
  • The proposed model outperforms several state-of-the-art deep learning architectures in driver distraction classification tasks.
  • The study validates the effectiveness of CBAM integration and data augmentation techniques for robust DMS in ADS.