Scholarlink

This paper presents a novel deep learning framework for pedestrian detection using Frequency-Modulated Continuous Wave (FMCW) radar, inspired by the MATLAB pedestrian detection example. Vision-based and LiDAR systems face challenges in adverse weather conditions, whereas FMCW radar offers a robust alternative. The proposed framework, based on deep learning, achieves a detection accuracy of 95%, computational efficiency improved by 30%, and enhanced robustness against cluttered conditions in urban environments. This marks significant advancements over existing pedestrian detection techniques in both accuracy and efficiency.

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

A Deep Learning Framework for Pedestrian Detection from FMCW Radar Data

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ayoub Fedjikh,
  • Dounia Dadghouj,
  • Mohammed Fattah,
  • Mohammed Mahfoudi,
  • Wafae El Hamdani,
  • Said Mazer,
  • Moulhime El Bakkali

摘要

This paper presents a novel deep learning framework for pedestrian detection using Frequency-Modulated Continuous Wave (FMCW) radar, inspired by the MATLAB pedestrian detection example. Vision-based and LiDAR systems face challenges in adverse weather conditions, whereas FMCW radar offers a robust alternative. The proposed framework, based on deep learning, achieves a detection accuracy of 95%, computational efficiency improved by 30%, and enhanced robustness against cluttered conditions in urban environments. This marks significant advancements over existing pedestrian detection techniques in both accuracy and efficiency.