[1]张善文,黄文准,张传雷.基于环境信息和深度自编码网络的农作物病害预测模型[J].江苏农业学报,2018,(02):288-292.[doi:doi:10.3969/j.issn.1000-4440.2018.02.009]
　ZHANG Shan-wen,HUANG Wen-zhun,ZHANG Chuan-lei.Forecasting model of crop disease based on environment information and deep auto-encoder network[J].,2018,(02):288-292.[doi:doi:10.3969/j.issn.1000-4440.2018.02.009]
点击复制

基于环境信息和深度自编码网络的农作物病害预测模型()
分享到：

江苏农业学报[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2018年02期
页码:
288-292
栏目:
植物保护
出版日期:
2018-04-25

文章信息/Info

Title:
Forecasting model of crop disease based on environment information and deep auto-encoder network
作者:
张善文1; 2; 黄文准1; 张传雷1
（1.西京学院信息工程学院，陕西西安710123；2.弗吉尼亚理工大学计算机科学系，美国弗吉尼亚VA24061）
Author(s):
ZHANG Shan-wen1; 2; HUANG Wen-zhun1; ZHANG Chuan-lei1
（1.College of Information Engineering, Xijing University, Xi’an 710123, China;2.Department of Computer Science, Virginia Tech, Blacksburg, VA24061, USA）
关键词:
病害预测; 环境信息; 自编码网络; 深度自编码网络
Keywords:
disease prediction; environment information; auto-encoder network; deep auto-encoder network
分类号:
TP391.4
DOI:
doi:10.3969/j.issn.1000-4440.2018.02.009
文献标志码:
A
摘要:
为了有效预测农作物病害，基于深度自编码网络，提出一种农作物病害预测模型。该模型能够自动从农作物环境信息中学习到主要的非线性组合特征，提高病害的预测精度。首先利用与农作物病害发生相关的环境信息构建病害预测的特征向量，并确定病害的4种预测状态，然后通过深度自编码网络从大量无标签的特征向量集中自动学习到可预测病害发生的深层特征的隐层参数，生成新特征向量集，再对有标签的新特征向量集进行学习，生成病害预测分类器，由此预测病害发生的等级。对黄瓜3种常见病害进行预测试验，平均预测准确率高达86%以上。试验结果表明，该模型是有效可行的，且具有较好的自学习更新能力。
Abstract:
In order to effectively forecast crop diseases, a forecasting model of crop disease was proposed based on deep auto-encode network. The model could automatically learn the senior nonlinear combination features from the original crop environment information data to improve the forecasting precision of the crop disease. The environment information data resulting in the crop disease were summarized and composed into the feature vector, and four kinds of disease prediction pattern were defined. Deep auto-encoder network was learned by the unlabeled feature vector set to build the hidden layer function and generate a new feature vector set, and the Softmax regression was used to classify the feature vectors based on the labeled feature set, and a classifier of disease prediction was obtained to forecast disease grade. Experiments on three common kinds of cucumber diseases were implemented. The average forecasting precision was over 86%. The experimental results indicated that the proposed method was effective and feasible, and had better automatically learning ability.

参考文献/References:

[1]PAVAN W, FRAISSE C W, PERES N A. Development of a web-based disease forecasting system for strawberries [J]. Computers & Electronics in Agriculture, 2011, 75(1):169-175.
[2]王翔宇,温皓杰,李鑫星,等. 农业主要病害检测与预警技术研究进展分析[J]. 农业机械学报, 2016, 47(9):266-277.
[3]李丽,李道亮,周志坚,等. 径向基函数网络与WebGIS融合的苹果病虫害预测[J]. 农业机械学报, 2008, 39(3):116-119.
[4]宋启堃,郑松,宋彦棠. 黔南州主要农作物病虫害监测预警专家系统的构建与实现[J]. 云南地理环境研究, 2011, 23(1):34-37.
[5]温皓杰,张领先,傅泽田,等. 基于Web的黄瓜病害诊断系统设计[J]. 农业机械学报, 2010, 41(12):178-182.
[6]曹志勇,邱靖,曹志娟,等.基于改进型神经网络的植物病虫害预警模型的构建[J].安徽农业科学, 2010,38(1):538-540.
[7]陈光绒,李小琴. 基于物联网技术的农作物病虫害自动测报系统[J]. 江苏农业科学, 2015, 43(4):406-410.
[8]姚卫平.贵池区小麦赤霉病发病程度中期预测模型[J].基层农技推广,2016,4(3):20-24.
[9]莫丽红. 基于APSO优化的小麦条锈病神经网络预测模型[J]. 湖北农业科学，2012，51（11）：2337-2340.
[10]王淑梅.气象条件与农作物病虫害预报和防治[J].世界农业,2010(2):55-57.
[11]王淑梅.基于气象视角的农作物病虫害预测预报研究概况[J].中国植保导刊, 2009(12):13-16.
[12]邓刚.气象因子的变化对黑龙江省森林病虫害影响的研究[D].哈尔滨:东北林业大学,2012.
[13]陈怀亮,张弘,李有.农作物病虫害发生发展气象条件及预报方法研究综述[J].中国农业气象, 2007,28(2):212-216
[14]马丽丽,纪建伟,贺超兴，等.番茄专家系统环境数据库在病害预测中的应用[J].农机化研究,2008(6):161-163.
[15]SHI M. Based on time series and RBF network plant disease forecasting[J]. Procedia Engineering, 2011, 15: 2384-2387.
[16]LECUN Y, BENGIO Y, HINTON G. Deep learning[J]. Nature, 2015, 521(7553): 436-444.
[17]谭娟,王胜春.基于深度学习的交通拥堵预测模型研究[J].计算机应用研究,2015, 32(10): 2951-2954.
[18]WANG Y, YOU Z, LI X, et al. Predicting protein－protein interactions from protein sequences by stacked sparse auto－encoder deep neural network [J]. Molecular BioSystems, 2017, 13(7):1336-1344.
[19]YOU Z H, ZHOU M C, LUO X, et al. Highly efficient framework for predicting interactions between proteins[J]. IEEE Transactions on Cybernetics, 2017, 47(3): 721-733.
[20]DYRMANN M, KARSTOFT H, MIDTIBY H S. Plant species classification using deep convolutional neural network[J]. Biosystems Engineering, 2016, 151: 72-80.
[21]SLADOJEVIC S, ARSENOVIC M, ANDERLA A, et al. Deep neural networks based recognition of plant diseases by leaf image classification[J]. Computational Intelligence and Neuroscience, 2016, 2016(6):1-11.
[22]黄冲,刘万才.试论物联网技术在农作物重大病虫害监测预警中的应用前景[J].中国植保导刊, 2015, 10:55-60.
[23]徐德良,王敏鑫,邵元.海物联网技术在茶树病虫害防治中的探讨[J]. 茶叶,2014，40(3):155-163.
[24]杨志民,梁静,刘广利.强模糊支持向量机在稻瘟病气象预警中的应用[J].中国农业大学学报, 2010,15(3):122-128.
[25]陈涛,高必梵,艾菊梅.基于BP神经网络的农作物虫害预测系统的研究[J].数字技术与应用, 2015(1):91-93.
[26]SHIN H C, ORTON M R, COLLINS D J, et al. Stacked auto-encoders for unsupervised feature learning and multiple organ detection in a pilot study using 4D patient data[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2013, 35(8): 1930-1943.

备注/Memo

备注/Memo:
收稿日期：2017-08-11 基金项目：国家自然科学基金项目(61473237) 作者简介：张善文(1965-)，男，陕西西安人，博士，教授，主要从事模式识别及其应用的研究。(E-mail)wjdw716@163.com 通讯作者：张传雷，(E-mail)a17647@gmail.com
更新日期/Last Update: 2018-05-04