面向采摘机器人的改进YOLOv3-tiny轻量化柑橘识别方法
DOI:
作者:
作者单位:

1.湖北工业大学农机工程研究设计院;2.湖北省农机装备智能化工程技术研究中心

作者简介:

通讯作者:

中图分类号:

TP391

基金项目:

国家重点研发计划项目(2017YFD0700603-03) 联合收割机复杂故障精准定位及远程诊断技术研究


Improved YOLOv3-tiny Lightweight Citrus Recognition Method for Picking Robot
Author:
Affiliation:

Institute of Agricultural Machinery, Hubei University of Technology

Fund Project:

National Key Research and Development Program of China (2017YFD0700603-03) Research on precise location and remote diagnosis technology of complex fault of combine Harvester

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

    针对柑橘采摘机器人快速、准确的识别需求,提出了一种基于改进的YOLOv3-tiny(You Only Look Once v3 tiny)的轻量化卷积神经网络模型的柑橘识别方法。为便于在算力有限的采摘机器人上应用,该方法用DIOU(Distance Intersection over Union)损失函数替换了YOLOv3-tiny卷积神经网络模型原有的损失函数,提高模型的识别定位精度;采用MobileNetv3-Small卷积神经网络模型替换了主干特征提取网络,使模型更加轻量化,提高模型的识别速度;在MobileNetv3-Small中加入了新的残差结构,减少主干网络特征信息的损失,进而提高模型的识别精度;在加强特征提取网络中加入了简化的空间金字塔池化SPP(Spatial Pyramid Pooling)网络结构和深度可分离卷积层集,提升模型提取特征信息的能力,再加入一个下采样层,将两个尺度间的特征信息充分融合,同时还加入了hard Swish激活函数,从而进一步提高模型的识别精度。通过与YOLOv3-tiny在柑橘测试集上的识别效果进行对比,改进的YOLOv3-tiny的平均识别精度mAP、F1值分别达到了96.52%、0.92,提高了3.24%、0.03,平均识别单幅图像所耗时间、模型权重大小仅为47 ms、16.9 MB,分别减少了24%、49%。通过与YOLOv3-tiny在针对柑橘测试集中处于不同环境条件下的柑橘的识别效果进行对比,改进的YOLOv3-tiny在光照充足且未遮挡条件下、光照充足且遮挡条件下、光照不足且未遮挡条件下、光照不足且遮挡条件下的柑橘正确识别率分别为98.6%、90.5%、95.8%、86.8%,分别提高了0.7%、6.5%、3.2%、7.7%。显示出改进YOLOv3-tiny轻量化柑橘识别方法具有识别精度高、识别速度快以及轻量化等特点。

    Abstract:

    In order to meet the rapid and accurate recognition requirements of citrus picking robot, a citrus recognition method based on improved YOLOv3-tiny(You Only Look Once v3 tiny) light convolutional neural network model was proposed. The proposed method replaces the original loss function of YOLOv3-tiny convolutional neural network model with DIOU(Distance Intersection over Union) loss function to improve the identification accuracy. Mobilenetv3-small convolutional neural network was used to replace the backbone feature extraction network, which made the model lighter and improved the recognition speed. A new residual structure is added in Mobilenetv3-small to reduce the loss of characteristic information and improve the identification accuracy. Simplified SPP(Spatial Pyramid Pooling) network structure and separable convolution layers are added to the backbone feature feature extraction network to improve the feature information extraction capability of the model. A down-sampling layer is used to integrate feature information of two scales, and the Hard Swish activation function is also applied. Proposed method was compared with the recognition effect of YOLOv3-tiny on citrus data set, the average mAP and F1 values of proposed method were 96.52% and 0.92, increased by 3.24% and 0.03, respectively. The recognition time cost and model weight were only 47 ms and 16.9 MB, respectively, which reduced by 24% and 49%. Further experiment was conducted to verify the recognition accuracy of citrus under different environmental conditions, The accuracy of YOLOv3-tiny were 98.6%, 90.5%, 95.8% and 86.8% respectively under sufficient light and no shade, insufficient light and no shade, and increased by 0.7%, 6.5%, 3.2% and 7.7%, respectively. The results showed that the improved YOLOv3-tiny light citrus recognition method achieves high recognition accuracy, fast recognition speed and light weight.

    参考文献
    相似文献
    引证文献
引用本文

汤旸,杨光友,王焱清. 面向采摘机器人的改进YOLOv3-tiny轻量化柑橘识别方法[J]. 科学技术与工程, , ():

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:2022-01-16
  • 最后修改日期:2022-04-22
  • 录用日期:2022-05-13
  • 在线发布日期:
  • 出版日期:
×
关于近期《科学技术与工程》编辑部居家办公的说明
亟待确认的版面费信息