Next-Generation Imaging Methodology: An Intelligent Transportation System for Consumer Industry

• Published in: IEEE transactions on consumer electronics Vol. 70; no. 1; pp. 3680 - 3687
• Main Authors: Devarajan, Ganesh Gopal, U, Kumaran, Chandran, Gopalakrishnan, Mahapatra, Rajendra Prasad, Alkhayyat, Ahmed
• Format: Journal Article
• Language: English
• Published: New York IEEE 01-02-2024; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accident detection; Adaptation models; Blockchain; blockchain technology; Blockchains; Consumer industry; Cryptography; Data models; Deep learning; Dynamics; Hybrid systems; Intelligent transportation systems; Intelligent vehicles; Internet of Vehicles; Rescue operations; Roads; Servers; Traffic accidents; Traffic congestion; Traffic models; Traffic safety; Transportation networks; Trusted third parties; Vehicle dynamics
• ISSN: 0098-3063; 1558-4127
• DOI: 10.1109/TCE.2024.3372906

Imaging technology is a recent advancement in developing smart vehicular environments in the consumer industry. Due to the increase in smart vehicles, road safety and other privacy issues become more common. This paper proposes a deep learning-based hybrid fatigue detection system where extracted features from ECG signals, the standard deviation of lane position, facial images, and detecting driver's drowsiness were combined using the Deep CNN network model. A sensor-based ResNet model is proposed for vehicle dynamics and road accident detection, which detects vehicle dynamics, road dynamics, and accidents and generates alarm signals for rescue operations. For traffic congestion detection and rerouting, we proposed a Blockchain-enabled Deep CNN network that detects traffic congestion using Deep CNN and gives optimal path suggestions using the Adaptive Cluster Head Selection (ACHS) approach. Furthermore, we introduced integrated blockchain technology with traffic congestion and routing models for enhancing security without depending on trusted third parties to run services. The proposed models performance is analyzed under different dynamics and compared with other state-of-the-art models.