Scholarlink

In this study, we extensively explored the most effective configurations of deep learning architectures, namely NanoDet and YOLOv8, for the real-time detection of orange fruits and assessment of their quality on mobile devices. Furthermore, we have successfully implemented a mobile application leveraging these deep learning models specifically tailored for the iOS platform. Our methodology involved the curation, annotation, and processing of a unique dataset comprising oranges. The empirical findings demonstrate that our system exhibits exceptional accuracy in both detecting orange fruits and evaluating their quality. The integration of artificial intelligence technologies presents a valuable tool for orange farms and consumers alike, facilitating rapid and precise assessments of orange fruits quality.

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 Approach for Real-Time Detection and Evaluation of Oranges on Mobile Devices

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Van Hong Son Than,
  • Dinh Minh Khoa Tran,
  • Thien Luong,
  • Duc Tho Le,
  • Viet Hung Le

摘要

In this study, we extensively explored the most effective configurations of deep learning architectures, namely NanoDet and YOLOv8, for the real-time detection of orange fruits and assessment of their quality on mobile devices. Furthermore, we have successfully implemented a mobile application leveraging these deep learning models specifically tailored for the iOS platform. Our methodology involved the curation, annotation, and processing of a unique dataset comprising oranges. The empirical findings demonstrate that our system exhibits exceptional accuracy in both detecting orange fruits and evaluating their quality. The integration of artificial intelligence technologies presents a valuable tool for orange farms and consumers alike, facilitating rapid and precise assessments of orange fruits quality.