Autonomous Systems in the Internet of Vehicles
by Balamurugan Balusamy, Sandeep Kumar Mathivanan, Prabhu Jayagopal, S. K. B. Sangeetha, Ali Kashif Bashir
2Adaptive Feature Alignment and Fusion for Multisensor Image Integration in the Internet of Vehicles
Vijay Anand R.1 and Madala Guru Brahmam2*
1School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, Tamilnadu, India
2School of Computer Science and Engineering, VIT-AP University, Amaravati, Andhra Pradesh, India
Abstract
Through the fusion of data from different sources, such as LiDAR, infrared, and RGB cameras, multisensor image fusion is the backbone of enhancing perception and decision-making for the Internet of Vehicles (IoV). To enhance real-time processing and resilience in IoV scenarios, the study proposes an advanced Adaptive Feature Alignment and Fusion Network (AFAF-Net). To achieve better feature stability and sensor disparity compensation, AFAF-Net conducts dynamic registration and multisensor image fusion through an attention-guided fusion mechanism. The KITTI and FLIR ADAS benchmark datasets of 14,999 multimodal image pairs captured in varied light and surroundings are used to test the solution presented here. Through experimental data, AFAF-Net provides real-time detection effectively at 97.3%, a 5.8% improved rate of object detection compared to existing solutions, and 23.6% reduced processing latency.
Keywords: Multisensor image fusion, Internet of Vehicles, LiDAR, RGB cameras, infrared, adaptive feature alignment, attention-guided fusion, real-time detection
2.1 Introduction
In medical and military applications ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access