AC-UNet: an improved UNet-based method for stem and leaf segmentation in Betula luminifera

被引:5
|
作者
Yi, Xiaomei [1 ]
Wang, Jiaoping [1 ]
Wu, Peng [1 ]
Wang, Guoying [1 ]
Mo, Lufeng [1 ]
Lou, Xiongwei [1 ]
Liang, Hao [1 ]
Huang, Huahong [1 ]
Lin, Erpei [1 ]
Maponde, Brian Tapiwanashe [1 ]
Lv, Chaihui [2 ]
机构
[1] Zhejiang A&F Univ, Coll Math & Comp Sci, Hangzhou, Peoples R China
[2] Hangzhou Ganzhi Technol Co Ltd, Hangzhou, Peoples R China
来源
关键词
Betula luminifera; stem and leaf division; UNET; hollow space pyramidal pooling; crossed attention;
D O I
10.3389/fpls.2023.1268098
中图分类号
Q94 [植物学];
学科分类号
071001 ;
摘要
Plant phenotypic traits play an important role in understanding plant growth dynamics and complex genetic traits. In phenotyping, the segmentation of plant organs, such as leaves and stems, helps in automatically monitoring growth and improving screening efficiency for large-scale genetic breeding. In this paper, we propose an AC-UNet stem and leaf segmentation algorithm based on an improved UNet. This algorithm aims to address the issues of feature edge information loss and sample breakage in the segmentation of plant organs, specifically in Betula luminifera. The method replaces the backbone feature extraction network of UNet with VGG16 to reduce the redundancy of network information. It adds a multi-scale mechanism in the splicing part, an optimized hollow space pyramid pooling module, and a cross-attention mechanism in the expanding network part at the output end to obtain deeper feature information. Additionally, Dice_Boundary is introduced as a loss function in the back-end of the algorithm to circumvent the sample distribution imbalance problem. The PSPNet model achieves mIoU of 58.76%, mPA of 73.24%, and Precision of 66.90%, the DeepLabV3 model achieves mIoU of 82.13%, mPA of 91.47%, and Precision of 87.73%, on the data set. The traditional UNet model achieves mIoU of 84.45%, mPA of 91.11%, and Precision of 90.63%, and the Swin-UNet model achieves . The mIoU is 79.02%, mPA is 85.99%, and Precision is 88.73%. The AC-UNet proposed in this article achieved excellent performance on the Swin-UNet dataset, with mIoU, mPA, and Precision of 87.50%, 92.71%, and 93.69% respectively, which are better than the selected PSPNet, DeepLabV3, traditional UNet, and Swin-UNet. Commonly used semantic segmentation algorithms. Experiments show that the algorithm in this paper can not only achieve efficient segmentation of the stem and leaves of Betula luminifera but also outperforms the existing state-of-the-art algorithms in terms of both speed. This can provide more accurate auxiliary support for the subsequent acquisition of plant phenotypic traits.
引用
收藏
页数:16
相关论文
共 50 条
  • [21] Medical image segmentation with UNet-based multi-scale context fusion
    Yuan, Yongqi
    Cheng, Yong
    SCIENTIFIC REPORTS, 2024, 14 (01):
  • [22] Soldering Defect Segmentation Method for PCB on Improved UNet
    Li, Zhongke
    Liu, Xiaofang
    APPLIED SCIENCES-BASEL, 2024, 14 (16):
  • [23] CellSegUNet: an improved deep segmentation model for the cell segmentation based on UNet++ and residual UNet models
    Sedat Metlek
    Neural Computing and Applications, 2024, 36 : 5799 - 5825
  • [24] Segmentation method of concrete CT pore and cracks based on improved UNet
    He J.
    Feng J.
    Jiao H.
    Li Y.
    Wu M.
    Han Y.
    Zhongguo Kuangye Daxue Xuebao/Journal of China University of Mining and Technology, 2023, 52 (03): : 615 - 624
  • [25] Glaucoma stage classification using UNET-based segmentation with multiple feature extraction technique
    Shyla, N. S. Jeya
    Emmanuel, W. R. Sam
    MULTIMEDIA TOOLS AND APPLICATIONS, 2024, 83 (30) : 74955 - 74971
  • [26] RAAWC-UNet: an apple leaf and disease segmentation method based on residual attention and atrous spatial pyramid pooling improved UNet with weight compression loss
    Wang, Jianlong
    Jia, Junhao
    Zhang, Yake
    Wang, Haotian
    Zhu, Shisong
    FRONTIERS IN PLANT SCIENCE, 2024, 15
  • [27] CellSegUNet: an improved deep segmentation model for the cell segmentation based on UNet plus plus and residual UNet models
    Metlek, Sedat
    NEURAL COMPUTING & APPLICATIONS, 2024, 36 (11): : 5799 - 5825
  • [28] Unfolded deep kernel estimation-attention UNet-based retinal image segmentation
    Radha, K.
    Yepuganti, Karuna
    Saritha, Saladi
    Kamireddy, Chinmayee
    Bavirisetti, Durga Prasad
    SCIENTIFIC REPORTS, 2023, 13 (01)
  • [29] RIC-Unet: An Improved Neural Network Based on Unet for Nuclei Segmentation in Histology Images
    Zeng, Zitao
    Xie, Weihao
    Zhang, Yunzhe
    Lu, Yao
    IEEE ACCESS, 2019, 7 : 21420 - 21428
  • [30] Unfolded deep kernel estimation-attention UNet-based retinal image segmentation
    K. Radha
    Karuna Yepuganti
    Saladi Saritha
    Chinmayee Kamireddy
    Durga Prasad Bavirisetti
    Scientific Reports, 13