无接触式猪体体尺测量技术研究进展
CSTR:
作者:
作者单位:

1.中山大学智能工程学院,深圳 518107;2.广西农垦永新畜牧有限公司,南宁 530000

作者简介:

李熙莹,E-mail:stslxy@mail.sysu.edu.cn

通讯作者:

中图分类号:

S817.3

基金项目:

国家生猪产业技术体系(CARS-35-PIG)


Progress on non-contact measurement of pig body size
Author:
Affiliation:

1.College of Intelligent Systems Engineering,Sun Yat-sen University,Shenzhen 518107,China;2.Guangxi Farms Yongxin Animal Husbandry Group Co.,Ltd.,Nanning 530000,China

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    随着视觉传感器和计算机视觉等领域的科技进步,无接触式猪体体尺检测相关的硬件设备和技术方法都发生了巨大的变化。视觉感知升级为三维立体视角,相应的设备也从黑白相机更新为深度设备。设备类型通常决定了基本的实施方案和技术方法,然而目前该领域的研究大多聚焦具体的算法细节,难以满足实际的工程需求。本文以视觉传感器类型为核心,概述了无接触式猪体体尺测量方面不同的工程部署方法,总结了设备使用方法、采集数据的环境条件和方式等,同时对猪体体尺测量的5个三维点云技术任务进行分析,并归纳了目前研究的优缺点、提升空间以及与实际工程应用相结合的方案,最后展望了深度学习与三维点云技术相结合的研究难点。通过对比发现,深度设备和三维点云技术是未来工程应用的方向,基于深度学习的点云分割、点云补全、关键点检测等技术具有良好的应用前景,为无接触式猪体体尺测量技术的后续研究提供了参考。

    Abstract:

    The hardware facilities and technical methods related to non-contact measurement of pig body size have undergone significant changes with the advancement of technology in fields including visual sensors and computer vision.Visual perception needs to meet three-dimensional requirements,and corresponding equipment have been updated from black and white cameras to depth equipment.The types of equipment usually determine the basic implementation plan and technical methods,however,current studies in this field mostly focus on specific algorithm details,which is difficult to meet the requirements of practical engineering.This article focused on the types of visual sensors and outlined different engineering deployment solutions for non-contact measurement of pig body size.The methods of equipment use and the environment and methods of data collection were summarized.The five tasks of 3D point cloud technology in measuring pig body size were analyzed.The advantages,disadvantages and potential for improvement of current studies and solutions for combination with applications of practical engineering were reviewed.The difficulties in studies on combining deep learning with 3D point cloud technology were prospected.The results of comparison showed that depth equipment and 3D point cloud technology are the directions for future applications of engineering in this field.Deep learning-based point cloud segmentation,point cloud completion,key point detection and other technologies have good prospects of application.It will provide a reference for the subsequent studies on non-contact measurement of pig body size.

    参考文献
    相似文献
    引证文献
引用本文

李熙莹,刘恒,苏淑妍,司景磊.无接触式猪体体尺测量技术研究进展[J].华中农业大学学报,2024,43(5):167-177

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2023-08-01
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2024-10-08
  • 出版日期:
文章二维码