Artificial intelligence analysis of contributive factors in determining blackleg disease severity in canola farmlands

被引:1
作者
Zhao, Liang [1 ]
Harding, Michael W. [2 ]
Peng, Gary [3 ]
Lange, Ralph [4 ]
Walkowiak, Sean [5 ]
Fernando, W. G. Dilantha [1 ]
机构
[1] Univ Manitoba, Dept Plant Sci, Winnipeg, MB R3T 2N2, Canada
[2] Alberta Agr Forestry & Rural Econ Dev, Crop Diversificat Ctr South, Brooks, AB T1R 1E6, Canada
[3] Agr & Agri Food Canada, Saskatoon Res & Dev Ctr, Saskatoon, SK 7N 0X2, Canada
[4] InnoTech Alberta Inc, Bioind Res Labs, Vegreville, AB T9C 1T4, Canada
[5] Canadian Grain Commiss, Grain Res Lab, Winnipeg, MB R3T 6C5, Canada
关键词
blackleg; crop rotation; deep learning; disease forecast; flea beetle; machine learning; altise; apprentissage automatique; apprentissage profond; prevision des maladies; necrose du collet; rotation des cultures; PHOMA STEM CANKER; OILSEED RAPE LEAVES; LEPTOSPHAERIA-MACULANS; FIELD CONDITIONS; INFECTION; EPIDEMIOLOGY; TEMPERATURE; ASCOSPORES; PYCNIDIOSPORES; CULTIVARS;
D O I
10.1080/07060661.2023.2290039
中图分类号
Q94 [植物学];
学科分类号
071001 ;
摘要
Canola (Brassica napus L.) production is threatened by blackleg disease caused by Leptosphaeria maculans. Disease outcome is determined by interactions among pathogens, plants, farming practices, and environmental factors. Although the gene-for-gene interactions between the pathogen and its plant host are relatively clear, how precisely the pathogen interacts with the environment and farming practices is still poorly understood, making disease forecasting challenging for commercial farmlands. In recent years, artificial intelligence (AI) has been successful in forecasting disease risks based on environmental factors. In this study, we evaluated two AI methods and a data augmentation method to forecast disease risk using a dataset collected from 116 farmlands in Alberta in 2021 and 2022. We first assessed a machine learning model (support vector machine or SVM) and a deep-learning model (convolutional neural network or CNN) to predict blackleg severity based on five weather variables, flea beetle damage, root maggot damage, and crop-rotation variables. Both SVM and CNN predicted the disease risk with an accuracy of over 66%. The data augmentation method did not improve model performance. Flea beetle feeding and maggot damage contribute little to the model's performance, and omitting these data did not appear to affect the results. In contrast, crop rotation contributes substantially to model performance. The five weather variables contribute roughly equally to the model's performance, and removing any of the individual weather variables did not impact prediction ability for both models. La production de canola (Brassica napus L.) est menacee par la necrose du collet causee par Leptosphaeria maculans. L'issue de la maladie est determinee par les interactions entre l'agent pathogene, la plante, les pratiques culturales et les facteurs environnementaux. Bien que les interactions gene pour gene entre l'agent pathogene et sa plante hote soient relativement claires, celles mettant precisement en cause l'environnement et les pratiques culturales ne sont pas bien comprises, ce qui constitue un defi quant a la prediction de la maladie sur les terres agricoles commerciales. Recemment, l'intelligence artificielle (IA) a reussi a predire les risques de maladie en se basant sur les facteurs environnementaux. Dans cette etude, nous avons evalue deux methodes d'IA et une technique d'augmentation des donnees pour predire le risque de maladie avec un jeu de donnees collectees, en 2021 et 2022, sur 116 terres agricoles de l'Alberta. Nous avons d'abord evalue un modele d'apprentissage automatique (machine a vecteur de support ou MVS) et un modele d'apprentissage profond (reseau neuronal a convolution ou RNC) pour predire la gravite de la necrose du collet en nous basant sur cinq variables climatiques, le dommage cause par l'altise, le dommage cause par la mouche des racines et les variables associees a la rotation des cultures. La SMV et le RNC ont predit le risque de maladie avec une precision de plus de 66%. La technique d'augmentation des donnees n'a pas ameliore la performance du modele. Les attaques d'altise et les dommages causes par la mouche des racines ont faiblement contribue a la performance du modele et l'omission de ces donnees ne semble pas avoir influence les resultats. En revanche, la rotation des cultures contribue substantiellement a la performance du modele. Les cinq variables climatiques contribuent a peu pres egalement a la performance du modele, et la suppression de quelque variable que ce soit n'a pas influence la capacite de prediction des deux modeles.
引用
收藏
页码:114 / 127
页数:14
相关论文
共 40 条
[1]  
BARBETTI M J, 1975, Australian Journal of Experimental Agriculture and Animal Husbandry, V15, P705, DOI 10.1071/EA9750705
[2]   Effects of temperature and wetness duration on infection of oilseed rape leaves by ascospores of Leptosphaeria maculans (stem canker) [J].
Biddulph, JE ;
Fitt, BDL ;
Leech, PK ;
Welham, SJ ;
Gladders, P .
EUROPEAN JOURNAL OF PLANT PATHOLOGY, 1999, 105 (08) :769-781
[3]  
Canola Council of Canada, 2020, FLEA BEETLES
[4]   Effects of seeding date and canola species on seedling damage by flea beetles in three ecoregions [J].
Carcamo, H. A. ;
Otani, J. K. ;
Dosdall, L. M. ;
Blackshaw, R. E. ;
Clayton, G. W. ;
Harker, K. N. ;
O'Donovan, J. T. ;
Entz, T. .
JOURNAL OF APPLIED ENTOMOLOGY, 2008, 132 (08) :623-631
[5]   Classification of hyperspectral remote-sensing data with primal SVM for small-sized training dataset problem [J].
Chi, Mingmin ;
Feng, Rui ;
Bruzzone, Lorenzo .
ADVANCES IN SPACE RESEARCH, 2008, 41 (11) :1793-1799
[6]   SUSCEPTIBILITIES OF SPECIES AND CULTIVARS AND CANOLA AND MUSTARD TO INFESTATION BY ROOT MAGGOTS (DELIA SPP) (DIPTERA, ANTHOMYIIDAE) [J].
DOSDALL, LM ;
HERBUT, MJ ;
COWLE, NT .
CANADIAN ENTOMOLOGIST, 1994, 126 (02) :251-260
[7]   Forecasting Plant and Crop Disease: An Explorative Study on Current Algorithms [J].
Fenu, Gianni ;
Malloci, Francesca Maridina .
BIG DATA AND COGNITIVE COMPUTING, 2021, 5 (01) :1-24
[8]  
Fernando WGD, 2018, PLANT DIS, V102, P790, DOI [10.1094/PDIS-05-17-0630-RE, 10.1094/pdis-05-17-0630-re]
[9]   World-wide importance of phoma stem canker (Leptosphaeria maculans and L-Biglobosa) on oilseed rape (Brassica napus) [J].
Fitt, BDL ;
Brun, H ;
Barbetti, MJ ;
Rimmer, SR .
EUROPEAN JOURNAL OF PLANT PATHOLOGY, 2006, 114 (01) :3-15
[10]   Comparison of disease severity and incidence at different growth stages of naturally infected canola plants under field conditions by pycnidiospores of Phoma lingam as a main source of inoculum [J].
Ghanbarnia, Kaveh ;
Fernando, W. G. Dilantha ;
Crow, Gary .
CANADIAN JOURNAL OF PLANT PATHOLOGY, 2011, 33 (03) :355-363