Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关实验视频

Updated: Jun 29, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

529

基于改进的深度双分辨率网络,用于识别异常排泄物的高效细分模型.

Pengguang He1,2,3, Rui Wu1,2,3, Da Liu1,3

  • 1College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, Zhejiang 310058, China.

Journal of animal science
|April 8, 2024
PubMed
概括

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Comparative Morphometric Study of the Osseous Labyrinth in Humans and Animal Models Using Centerline Analysis.

Noise & health·2026
Same author

Ufl1-Mediated UFMylation Sustains Amelogenesis by Stabilizing RUNX2.

FASEB journal : official publication of the Federation of American Societies for Experimental Biology·2026
Same author

Comparative effectiveness and safety of ondansetron-based dual versus triple antiemetic regimens for chemotherapy-induced nausea and vomiting.

Medicine·2026
Same author

Elevating Tetracycline Removal in Microbial Electrochemical Systems: Insights from Extracellular Electron Transfer and Rational Aggregation of Microbial Consortia.

ACS applied materials & interfaces·2026
Same author

Biological applications and future perspectives of bioactive compounds as pig feed additives.

Animal nutrition (Zhongguo xu mu shou yi xue hui)·2026
Same author

Aging of mesenchymal stem cells in bone aging: Mechanisms, impact, and therapeutic perspectives.

Ageing research reviews·2026

这项研究开发了一种轻量级细分模型 (DDRNet-s-KD),用于自动监测异常的排泄物. 该模型实现了高精度和速度,使其适用于农场的边缘设备.

科学领域:

  • 农业工程 农业工程
  • 计算机视觉 计算机视觉
  • 机器学习 机器学习

背景情况:

  • 排便的特征表明群体的健康状况.
  • 以前用于便的物体检测方法面临着标签和准确性挑战.
  • 农场边缘计算需要高效的实时算法.

研究的目的:

  • 开发一种轻量级的细分模型,用于自动化监测异常排泄物.
  • 优化模型,以便在资源受限的边缘设备上部署.
  • 与现有方法相比,提高准确性和推断速度.

主要方法:

  • 重新定义了排泄物识别作为细分任务.
  • 增强了带有注意力机制和多损失功能的基础细分网络 (DDRNet).
  • 开发了一个轻量级版本 (DDRNet-s-KD),使用组卷积和知识蒸.
  • 将模型量化为8位整数,并转换为TensorRT格式用于边缘部署.

主要成果:

  • DDRNet-s-KD模型实现了79.43%和86.10 FPS的平均子系数 (mDice).
  • 与基准相比,mDice增长了2.91%和FPS增长了61.2%.
  • 量子化模型大小减少到13.7 MB (82.96%的减少),并在Jetson Xavier NX上实现了137.51 FPS.
关键词:
边缘设备 边缘设备轻量级的轻量级的轻量级的轻量级的一个模型的压缩压缩.语义细分 语义细分 语义细分 语义细分

更多相关视频

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.1K
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K

相关实验视频

Last Updated: Jun 29, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

529
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.1K
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

2.5K

结论:

  • 拟议的轻量级细分模型对于监测异常的排泄物是有效的.
  • 该模型的效率和小尺寸适合在农业环境中部署边缘设备.
  • 这种方法为其他农业嵌入式系统提供了有价值的参考.