基于边界样本分位数的葡萄霜霉病自适应识别方法

doi:10.13733/j.jcam.issn.2095-5553.2024.10.036

中国农机化学报 ›› 2024, Vol. 45 ›› Issue (10): 247-253.DOI: 10.13733/j.jcam.issn.2095-5553.2024.10.036

基于边界样本分位数的葡萄霜霉病自适应识别方法

高芮1,2，武琼2，韩玮1，杨涛2，卢晨媛3，张黎4

(1. 南京信息工程大学应用气象学院，南京市，210044； 2. 旬邑县气象局，陕西咸阳，711300；
3. 永寿县气象局，陕西咸阳，713400； 4. 陕西省气象台，西安市，710014)

出版日期:2024-10-15 发布日期:2024-09-30
基金资助:
国家自然科学基金青年科学基金(42005140)

Adaptive identification method of grape downy mildew based on quantile of boundary sample

Gao Rui1, 2, Wu Qiong2, Han Wei1, Yang Tao2, Lu Chenyuan3, Zhang Li4

(1. College of Applied Meteorology, Nanjing University of Information Engineering, Nanjing, 210044, China;
2. Xunyi County Meteorological Bureau, Xianyang, 711300, China; 3. Yongshou County Meteorological Bureau,
Xianyang, 713400, China; 4. Shaanxi Meteorological Observatory, Xi'an, 710014, China)

Online:2024-10-15 Published:2024-09-30

摘要/Abstract

摘要： 针对葡萄霜霉病病斑组织图像阈值难以确定的问题，提出一种基于边界样本分位数的自适应阈值确定方法，通过高斯滤波识别病斑边界，并采用边界样本的50%分位数确定为病斑阈值。之后采用蒙特卡洛方法，通过随机采样方法估算病斑比例。结果表明，与其他阈值确定方法对比，所提方法能够自适应获取病斑灰度阈值，识别精度达到92.2%，明显高于其他阈值确定方法；与传统的机器学习方法对比，在识别精度上高于BP神经网络、卷积神经网络、支持向量机，略低于VGG16模型的94.3%与ResNet50模型的96.26%，但计算时间为1.410 s，远快于VGG16模型与ResNet50模型的5.588 s与20.317 s，说明方法能够在较短的运行时间内实现较高的精度。

关键词: 葡萄霜霉病, 边界识别, 样本分位数, 病斑识别, 高斯滤波

Abstract: Aiming at the problem of difficulty to determine the threshold of grape downy mildew lesion tissue image, an adaptive threshold determination method based on quantiles of boundary samples was proposed. The boundary of diseased spot was identified by Gaussian filtering, and the threshold of diseased spot was determined by 50% quantiles of boundary sample. Then Monte Carlo method was used to estimate the proportion of diseased spots by random sampling method. The results show that compared with other threshold determination methods, the proposed method can adaptively obtain the grayscale threshold of the lesion, with a recognition accuracy of 92.2%, which is significantly higher than other threshold determination methods. Compared with the traditional machine learning methods, the recognition accuracy of this method is higher than that of BP neural network, convolutional neural network, and support vector machine, slightly lower than 94.3% of the VGG16 model and 96.26% of the ResNet50 model. However, the calculation time is 1.410 s, which is much faster than 5.588 s and 20.317 s of the VGG16 model and ResNet50 model, indicating that this method can achieve high accuracy in a shorter running time.

Key words: grape downy mildew, boundary recognition, sample quantile, plaque recognition, Gaussian filter

中图分类号:

S436.631.1

高芮, 武琼, 韩玮, 杨涛, 卢晨媛, 张黎. 基于边界样本分位数的葡萄霜霉病自适应识别方法[J]. 中国农机化学报, 2024, 45(10): 247-253.

Gao Rui, , Wu Qiong, Han Wei, Yang Tao, Lu Chenyuan, Zhang Li. Adaptive identification method of grape downy mildew based on quantile of boundary sample[J]. Journal of Chinese Agricultural Mechanization, 2024, 45(10): 247-253.

参考文献

[ 1 ] 杜娟, 李金, 张涛, 等. 应用real‑time PCR定量检测果园葡萄霜霉病菌潜伏侵染[J]. 江苏农业学报, 2021, 37(4): 861-866.
Du Juan, Li Jin, Zhang Tao, et al. Quantitative detection of latent infection of grape downy mildew in field by real‑time PCR [J]. Jiangsu Journal of Agricultural Sciences, 2021, 37(4): 861-866.
[ 2 ] 杨付前, 马靖艳. 葡萄霜霉病发生规律及防治措施[J]. 林业与生态, 2021(5): 42.
[ 3 ] El‑Helly M, El‑Beltagy S, Rafea A. Image analysis based interface for diagnostic expert systems [C]. Proceedings of the Winter International Synposium on Information and Communication Technologies, Trinity College Dublin, 2004: 1-6.
[ 4 ] Sammany M, Medhat T. Dimensionality reduction using rough set approach for two neural networks‑based applications [A]. Rough Sets and Intelligent Systems Paradigms [M]. Heidelberg: Springer Berlin, 2007.
[ 5 ] 张建华, 孔繁涛, 吴建寨, 等. 基于改进VGG卷积神经网络的棉花病害识别模型[J]. 中国农业大学学报, 2018, 23(11): 161-171.
Zhang Jianhua, Kong Fantao, Wu Jianzhai, et al. Cotton disease identification model based on improved VGG convolution neural network [J]. Journal of China Agricultural University, 2018, 23(11): 161-171.
[ 6 ] 杨森, 冯全, 张建华. 基于深度学习与复合字典的马铃薯病害识别方法[J]. 农业机械学报, 2020, 51(7): 22-29.
Yang Sen, Feng Quan, Zhang Jianhua. Identification method for potato disease based on deep learning and composite dictionary [J]. Transactions of the Chinese Society for Agricultural Machinery, 2020, 51(7): 22-29.
[ 7 ] Wei X K, Yang Z M, Liu Y X, et al. Railway track fastener defect detection based on image processing and deep learning techniques: A comparative study [J]. Engineering Applications of Artificial Intelligence, 2019, 80(4): 66-80.
[ 8 ] 骆润玫, 王卫星. 基于卷积神经网络的植物病虫害识别研究综述[J]. 自动化与信息工程, 2021, 42(5): 1-10.
Luo Runmei, Wang Weixing. Review on plant disease and pest identification based on convolutional neural network [J]. Automation and Information Engineering, 2021, 42(5): 1-10.
[ 9 ] 杨红云, 肖小梅, 黄琼, 等. 基于卷积神经网络和迁移学习的水稻害虫识别[J]. 激光与光电子学进展, 2022, 59(16): 323-330.
Yang Hongyun, Xiao Xiaomei, Huang Qiong, et al. Rice pest identification based on convolutional neural network and transfer learning [J]. Laser & Optoelectronics Progress, 2022, 59(16): 323-330.
[10] Wu Y, Lu Y J. An intelligent machine vision system for detecting surface defects on packing boxes based on support vector machine [J]. Measurement and Control, 2019, 52(7): 1102-1110.
[11] 吕泽卿, 付兴建. 基于改进Canny算法的绝缘子裂缝识别[J]. 北京信息科技大学学报(自然科学), 2020, 35(6): 25-30.
Lü Zeqing, Fu Xingjian. Insulator crack recognition based on improved Canny algorithm [J]. Journal of Beijing Information Science & Technology University, 2020, 35(6): 25-30.
[12] 袁向荣. 边缘识别的二维正交多项式拟合及结构变形检测[J]. 图学学报, 2014, 35(1): 79-84.
Yuan Xiangrong. Orthogonal polynomial fitting for the edge detection and beam deformation measurement [J]. Journal of Graphics, 2014, 35(1): 79-84.
[13] 潘晋新, 景博, 焦晓璇, 等. 多应力耦合条件下氧气浓缩器退化建模[J]. 北京航空航天大学学报, 2023, 49(2): 472-481.
Pan Jinxin, Jing Bo, Jiao Xiaoxuan, et al. Degradation modeling of oxygen concentrator in multiple stress coupling [J]. Journal of Beijing University of Aeronautics and Astronautics, 2023, 49(2): 472-481.
[14] 淮妮. 基于蒙特卡洛的输送机电液伺服系统动态特性分析[J]. 国外电子测量技术, 2021, 40(4): 6-9.
[15] 王卫东, 雷晓鸣, 杜香刚, 等. 基于蒙特卡洛和BIM的CRTSⅢ型板式无砟轨道铺设工期仿真[J]. 中南大学学报(自然科学版), 2019, 50(7): 1655-1661.
Wang Weidong, Lei Xiaoming, Du Xianggang, et al. Period‑simulation of CRTS Ⅲ slab track based on Monte Carlo simulation and BIM [J]. Journal of Central South University (Science and Technology), 2019, 50(7): 1655-1661.
[16] Alaya M B, Kebaier A. Central limit theorem for the multilevel Monte Carlo Euler method [J]. The Annals of Applied Probability, 2015, 25(1): 1050-5164.

基于边界样本分位数的葡萄霜霉病自适应识别方法

Adaptive identification method of grape downy mildew based on quantile of boundary sample

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 1

编辑推荐

Metrics

本文评价