Scholarlink

In order to develop intelligent transportation systems, especially for applications like driver-assist technologies and autonomous vehicles, traffic sign recognition is essential. In this chapter, we present a CNN-based real-time system that is lightweight and was constructed using the Keras framework. To make our model more regionally adaptive, we trained it on traffic sign datasets from Saudi Arabia, Tunisia, Germany, and Italy. In order to ensure that the system can function in real-world settings with limited hardware resources, we also optimized it for edge devices like the Raspberry Pi and Jetson Nano. The system’s architecture, training process, data preprocessing, and final deployment are all described in this paper. Standard metrics and real-time testing were used for evaluation in order to verify its functionality.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Quantum-Inspired Deep Neural Network for Real-Time Traffic Sign Recognition in 6G-Enabled Intelligent Transportation Systems

  • 章节

  • 发表日期：

    2026年

查看全文
  • Vishal Verma,
  • Tarang Verma,
  • Arjun Sharma,
  • Sanna Mehraj Kak,
  • Manish Chaudhary,
  • Hisham Mohammad

摘要

In order to develop intelligent transportation systems, especially for applications like driver-assist technologies and autonomous vehicles, traffic sign recognition is essential. In this chapter, we present a CNN-based real-time system that is lightweight and was constructed using the Keras framework. To make our model more regionally adaptive, we trained it on traffic sign datasets from Saudi Arabia, Tunisia, Germany, and Italy. In order to ensure that the system can function in real-world settings with limited hardware resources, we also optimized it for edge devices like the Raspberry Pi and Jetson Nano. The system’s architecture, training process, data preprocessing, and final deployment are all described in this paper. Standard metrics and real-time testing were used for evaluation in order to verify its functionality.