Instance Segmentation Model for Microscopic Image of Citrus Main Leaf Vein Based on Mask R-CNN

被引：0
|
作者：
Weng H. [1 ,2 ]
Li X. [1 ,2 ]
Xiao K. [1 ,2 ]
Ding R. [1 ,2 ]
Jia L. [3 ]
Ye D. [1 ,2 ]
机构：
[1] College of Mechanical and Electrical Engineering, Fujian Agriculture and Forestry University, Fuzhou
[2] Fujian Key Laboratory of Agricultural Information Sensing Technology, Fuzhou
[3] School of Information Engineering, Huzhou University, Huzhou
来源：
Nongye Jixie Xuebao/Transactions of the Chinese Society for Agricultural Machinery | 2023年 / 54卷 / 07期
关键词：
citrus main leaf vein; instance segmentation; mask region convolutional neural network; microscopic image; microscopic phenotypes;
D O I：
10.6041/j.issn.1000-1298.2023.07.025
中图分类号：
学科分类号：
摘要：
There is a low efficiency of automatically measuring and analyzing plant anatomic phenotypes currently, which makes it difficult to well deal with the issue of extracting and recognizing the complex anatomical phenotypes. In order to solve this problem, a mask region convolutional neural network (Mask R - CNN) based instance segmentation model for microscopic images of the citrus main leaf veins was proposed. In this model, the deep residual network (ResNet50) and the feature pyramid network (FPN) were used as the backbone feature extraction network. In addition, a new region of interest Align (Rol-Align) layer was added to the Mask branch to improve the segmentation accuracy. The results showed that the network can accurately identify and segment pith, xylem, phloem and cortical cells, respectively, in the citrus main leaf veins. The average precision (IoU was 0.50) of the model for segmentation of pith, xylem, phloem and cortical cells was 98.9%, 89.8%, 95.7% and 97.2%, respectively, and the overall average precision (IoU was 0.50) for segmentation of the four tissue regions was 95. 4%. The mean average precision of Mask R - CNN with adding Rol - Align to the Mask branch was improved by 1. 6 percentage points compared with that without. The results showed that Mask R - CNN model presented good performance of recognition and segmentation of various tissue regions of citrus main leaf veins, which can provide technical support for citrus microscopic phenotyping. © 2023 Chinese Society of Agricultural Machinery. All rights reserved.
引用
收藏
页码：252 / 258and271
相关论文
共 36 条
  • [1] PIERUSCHKA R, SCHURR U., Plant phenotyping
  • [2] past, present, and future, Plant Phenomies, 3, pp. 1-6, (2019)
  • [3] JEZ J M, TOPP G N, BUCKNER E, Et al., High-throughput image segmentation and machine learning approaches in the plant sciences across multiple scales, Emerging Topics in Life Sciences, 5, 2, pp. 239-248, (2021)
  • [4] DENG Hanbing, XU Tongyu, ZHOU Yuneheng, Et al., Segmentation model for maize plant images based on depth mask, Transactions of the CSAE, 37, 18, pp. 109-120, (2021)
  • [5] DENG Ying, WU Huarui, ZHU Huaji, Recognition and counting of citrus flowers based on instance segmentation, Transactions of the CSAE, 36, 7, pp. 200-207, (2020)
  • [6] YUAN Shan, TANJ Hao, GUO Ya, Segmentation method for plant leaves using an improved Mask R - CNN model, Transactions of the CSAE, 38, 1, pp. 212-220, (2022)
  • [7] FU L, WANG B., CUI Y, Et al., Kiwifruit recognition at nighttime using artificial lighting based on machine vision [J], International Journal of Agricultural and Biological Engineering, 4, pp. 52-59, (2015)
  • [8] STROCK C F, SCHNEIDER H M, LYNCH J P., Anatomies
  • [9] phenotyping of plant anatomy, Trends in Plant Science, 27, 6, pp. 520-523, (2022)
  • [10] WANG Shiya, ZHENG Dianfeng, FENG Naijie, Et al., Effects of uniconazole on physiological characteristics and microstructure under waterlogging stress at seedling stage in soybean[J], Acta Agronomica Sinica, 47, 10, pp. 1988-2000, (2021)
1234