Scholarlink

To overcome the limitations of previous R-CNN and YOLO models, researchers have introduced improved algorithms that enhance accuracy and detection speed. This paper examines two such models: YOLOv5 from the YOLO family and Faster R-CNN from the R-CNN family. Utilizing the S2TLD dataset and the PyTorch framework, we implement and compare these models to assess their strengths, weaknesses, and advancements over prior versions. Experimental results indicate that the proposed model demonstrates higher reliability than the YOLOv5m model, achieving a recall of 98.2%.

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

Developing an Object Detection Algorithm in Traffic Light Signal

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Trinh Nguyen Thi,
  • Tinh Tran Van,
  • Duc Vu Xuan,
  • Phuc Dao Xuan,
  • Thao Dao Le Thu,
  • Thanh Le Thi Hai,
  • Dung Nguyen Tien,
  • Phat Nguyen Huu

摘要

To overcome the limitations of previous R-CNN and YOLO models, researchers have introduced improved algorithms that enhance accuracy and detection speed. This paper examines two such models: YOLOv5 from the YOLO family and Faster R-CNN from the R-CNN family. Utilizing the S2TLD dataset and the PyTorch framework, we implement and compare these models to assess their strengths, weaknesses, and advancements over prior versions. Experimental results indicate that the proposed model demonstrates higher reliability than the YOLOv5m model, achieving a recall of 98.2%.