Abstract:
Distracted driving is a major cause of road accidents around the world. To reducesuch incidents, we need smarter and faster systems that can detect distractions inreal time. This study presents a hybrid model that combines quantum computingwith deep learning to detect distracted driving. The proposed quantum inspireddeep learning model uses the power of quantum computing to process highdimensionaldata more efficiently. It also takes advantage of classical deeplearning methods for strong pattern recognition. First, images are preprocessedby resizing, augmenting, and normalizing them. Then, they are converted into212-dimensional vectors for quantum processing. The quantum part of themodel uses 12 qubits and applies 10 layers of entangled quantum gates. Afterprocessing, it uses Pauli-Z gates to extract important features. These features arethen passed to a SoftMax classifier, which predicts the driver’s behavior. Twopopular datasets, State Farm Distracted Driver Detection and AUC DistractedDriver, were used for training and testing of the proposed model. Both datasetscontain 10 types of driving behaviors, such as texting, talking, or eating whiledriving. The proposed model achieved a high accuracy of 99.11%, performingbetter than standard CNN models.
Page(s):
119-119
DOI:
DOI not available
Published:
Journal: 4th International Conference of Sciences “Revamped Scientific Outlook of 21st Century, 2025” , November 12,2025, Volume: 1, Issue: 1, Year: 2025
Keywords:
quantum inspired deep learning framework
,
distracted driving behavior
,
Driver attention monitoring