Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Jul 4, 2026

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

LiteDriveNet for driver distraction classification using a lightweight multi-scale convolutional neural network.

Abirami Namachivayam¹, Nirmala Paramanandham²

¹School of Electronics Engineering, Vellore Institute of Technology, Chennai, Tamil Nadu, 600 127, India.

Scientific Reports

|July 2, 2026

Summary

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

Deep intelligence: a four-stage deep network for accurate brain tumor segmentation.

Scientific reports·2025

Same author

Pixel Intensity Resemblance Measurement and Deep Learning Based Computer Vision Model for Crack Detection and Analysis.

Sensors (Basel, Switzerland)·2023

See all related articles

A new lightweight convolutional neural network, LiteDriveNet, effectively identifies distracted driving behaviors. This efficient model offers improved accuracy and computational performance for real-time road safety applications.

Area of Science:

Computer Vision
Artificial Intelligence
Road Safety

Background:

Distracted driving is a significant global safety issue contributing to road accidents.
Existing models often lack the efficiency required for real-time deployment.

Purpose of the Study:

Introduce LiteDriveNet, a novel, resource-efficient convolutional neural network for distracted driving detection.
Evaluate LiteDriveNet's performance against state-of-the-art models on diverse datasets.

Main Methods:

Developed LiteDriveNet, a 16-layer CNN (1.71 MB) with multi-scale receptive blending and hierarchical feature aggregation.
Created DistractedDrivingSet_v1 (6075 images, 8 classes) and utilized SFD3 and AUCv2 benchmark datasets.
Performed comparative analysis and ablation studies to validate model efficacy.

Keywords:

Advanced driver assistance system Artificial Intelligence Convolutional neural networks Distracted driver classification

Related Experiment Videos

Last Updated: Jul 4, 2026

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

End-To-End Deep Neural Network for Salient Object Detection in Complex Environments

Published on: December 15, 2023

Main Results:

LiteDriveNet demonstrated superior accuracy and computational efficiency across all tested datasets.
Achieved accuracy improvements of 16.53% (DistractedDrivingSet_v1), 6.70% (SFD3), 10.91% (AUCv2 Camera1), and 28.97% (AUCv2 Camera2).
Outperformed lightweight and neural architecture search-based models.

Conclusions:

LiteDriveNet is a highly effective lightweight model for real-time distracted driver recognition.
The model's efficiency and accuracy make it suitable for practical road safety systems.
This research contributes a novel architecture and dataset for advancing distracted driving detection.