Open Access Open Access  Restricted Access Subscription Access

VISIONGUARD- Object Detection in Adverse Weather Conditions with Health Monitoring

Dr. Ashwani Sharma, Rohit J Sangan, Nitinkumar Loni, Yashwanth P, Thoshith Kumar B, Amruth N Murthy, Kavan N Murthy

Abstract


Driving in bad weather like rain, fog, or low light often makes it hard for traditional vehicle detection systems to work effectively. To tackle this, we introduce VisionGuard—a smart solution that combines advanced object detection with real-time driver health and drowsiness monitoring. Powered by YOLOv8 and trained with the ACDC dataset, it’s built to handle tough visual conditions. It also uses ultrasonic and infrared sensors to improve how it sees the environment. On the inside, it tracks vital signs like heart rate, oxygen levels, and body temperature through wearable sensors, while keeping an eye on signs of fatigue using facial landmarks and blink detection. With over 95% detection accuracy, VisionGuard brings together deep learning, sensor fusion, and health monitoring into one system—making roads safer by looking out for both what’s outside the vehicle and how the driver is doing.


Full Text:

PDF

References


Garg, K. et al.,” Synthetic Adverse Weather Data for Object Detection,” Journal of Computer Vision, 2021.

Bochkovskiy, A., Wang, C.-Y., Liao, H.-Y. M., “YOLOv4: Optimal Speed and Accuracy of Object Detection,” arXiv preprint arXiv, 2004.10934, 2020.

Redmon, J. et al., “You Only Look Once: Unified, Real- Time Object Detection” CVPR, 2016.

Simonyan, K., Zisserman, A.,”Very Deep Convolutional Net- works for Large-Scale Image Recognition,” arXiv preprint arXiv:, arXiv:1409.1556, 2014.

Zeiler, M. D., Fergus, R., “Visualizing and Understanding Convolutional Networks,” ECCV, 2014.

Huang, G. et al., “Densely Connected Convolutional Networks,”CVPR2017.

Sandler, M. et al., “MobileNetV2: Inverted Residuals and Linear Bot- tlenecks,” CVPR2018.

Zhou, B. et al., “Learning Deep Features for Discriminative Localiza- tion,” IEEE Internet Computing, vol. 22, no. 6, pp. 36-43, 1 Nov.-Dec. 2018, doi: 10.1109/MIC.2018.252095348.

Dosovitskiy, A. et al., “An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale,” arXiv preprint arXiv, 2010.11929, 2020.

Tan, M., Le, Q. V., “EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks,” ICML2019.


Refbacks

  • There are currently no refbacks.