Scholarlink

This research investigates the integration of cutting- edge deep learning models to create a robust, high- performance system for dog breed identification. The project utilizes YOLO (You Only Look Once) for rapid and precise object detection, enabling the identification and localization of dogs within images with exceptional speed. Complementing this, EfficientNet-B0 is employed for detailed breed classification, leveraging its scalable architecture to capture subtle differences between breeds with high accuracy. By combining YOLO’s strength in real-time detection with EfficientNet-B0’s fine-grained classification capabilities, this approach addresses the challenges of both speed and accuracy in breed recognition. The project thoroughly examines the models’ comparative strengths and limitations, focusing on metrics such as accuracy, computational efficiency, and data requirements, ultimately delivering an advanced and practical solution for real- world applications in dog breed identification.

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

Scalable Dog Breed Classification with YOLO and EfficientNet-B0: Enhancing Accuracy and Speed Through High-Performance Computing

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • K. Gowtham Natraj,
  • S. Tharun,
  • Mohith Niranjen,
  • Kaushik Karthikeyan

摘要

This research investigates the integration of cutting- edge deep learning models to create a robust, high- performance system for dog breed identification. The project utilizes YOLO (You Only Look Once) for rapid and precise object detection, enabling the identification and localization of dogs within images with exceptional speed. Complementing this, EfficientNet-B0 is employed for detailed breed classification, leveraging its scalable architecture to capture subtle differences between breeds with high accuracy. By combining YOLO’s strength in real-time detection with EfficientNet-B0’s fine-grained classification capabilities, this approach addresses the challenges of both speed and accuracy in breed recognition. The project thoroughly examines the models’ comparative strengths and limitations, focusing on metrics such as accuracy, computational efficiency, and data requirements, ultimately delivering an advanced and practical solution for real- world applications in dog breed identification.