Prediction of the apple scab using machine learning and simple weather stations

Abstract Apple scab is an economically important pest for apple production. It is controlled by applying fungicides when conditions are ripe for the development of its spores. This occurs when leaves are wet for a long enough time at a given temperature. However, leaf wetness is not a sufficiently well-defined agro-meteorological variable. Moreover, the readings of leaf wetness sensors depend to a large extent on their location within the tree canopy. Here we show that virtual wetness sensors, which are based on the easily obtained meteorological parameters such as temperature, relative humidity and wind speed, can be used in place of physical sensors. To this end, we have first collected data for two growing seasons from two types of wetness sensors planted in four locations in the tree canopy. Then, for each sensor we have built a machine-learning model of leaf wetness using the aforementioned meteorological variables. These models were further used as virtual sensors. Finally, Mills models of apple scab infection were built using both real and virtual sensors and their results were compared. The comparison of apple scab models based on real sensors shows significant variability. In particular, the results of a model depend on the location of the sensor within the canopy. The models based on data obtained from virtual sensors, are similar to the models based on physical sensors. Both types of models generate results within the same range of variability. The outcome of the study shows that the control of apple scab can be based on machine learning models based on standard meteorological variables. These variables can be readily obtained using inexpensive meteorological stations equipped with basic sensors. These results open the way to a widespread application of precise control of apple scab and consequently significant reduction of the use of pesticides in apple production with benefits for environment, human health and economics of production.

[1]  Rich Caruana,et al.  An empirical comparison of supervised learning algorithms , 2006, ICML.

[2]  Bartek Wilczynski,et al.  Active enhancer positions can be accurately predicted from chromatin marks and collective sequence motif data , 2013, BMC Systems Biology.

[3]  Yong Peng,et al.  Modeling reference evapotranspiration using extreme learning machine and generalized regression neural network only with temperature data , 2017, Comput. Electron. Agric..

[4]  Andy Liaw,et al.  Classification and Regression by randomForest , 2007 .

[5]  R Core Team,et al.  R: A language and environment for statistical computing. , 2014 .

[6]  Elpiniki I. Papageorgiou,et al.  Hybrid learning of fuzzy cognitive maps for sugarcane yield classification , 2016, Comput. Electron. Agric..

[7]  L. Cang,et al.  Microbial and enzyme properties of apple orchard soil as affected by long-term application of copper fungicide , 2009 .

[8]  Alejandro Castaeda-Miranda,et al.  Smart frost control in greenhouses by neural networks models , 2017 .

[9]  Witold R. Rudnicki,et al.  Feature Selection with the Boruta Package , 2010 .

[10]  Rafał Niemiec,et al.  Application of all-relevant feature selection for the failure analysis of parameter-induced simulation crashes in climate models , 2015 .

[11]  H. B. Mann,et al.  On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other , 1947 .

[12]  L. Baum,et al.  Statistical Inference for Probabilistic Functions of Finite State Markov Chains , 1966 .

[13]  Herb S. Aldwinckle,et al.  compendium of apple and pear diseases , 1990 .

[14]  Leo Breiman,et al.  Random Forests , 2001, Machine Learning.

[15]  Reza Ehsani,et al.  Field detection of anthracnose crown rot in strawberry using spectroscopy technology , 2017, Comput. Electron. Agric..

[16]  Karl Pearson F.R.S..  LIII. On lines and planes of closest fit to systems of points in space , 1901 .

[17]  Robert J. McQueen,et al.  Applying machine learning to agricultural data , 1995 .

[18]  Roger D. Magarey,et al.  Grape canopy surface wetness: Simulation versus visualization and measurement , 2006 .

[19]  M. Saudreau,et al.  Comparison of Penman–Monteith and non-linear energy balance approaches for estimating leaf wetness duration and apple scab infection , 2011 .

[20]  Senén Barro,et al.  Do we need hundreds of classifiers to solve real world classification problems? , 2014, J. Mach. Learn. Res..

[21]  Wei-Yin Loh,et al.  Classification and regression trees , 2011, WIREs Data Mining Knowl. Discov..

[22]  Gene H. Golub,et al.  Singular value decomposition and least squares solutions , 1970, Milestones in Matrix Computation.

[23]  F. Wilcoxon.  Individual Comparisons by Ranking Methods , 1945 .

[24]  Rich Caruana,et al.  An empirical evaluation of supervised learning in high dimensions , 2008, ICML '08.

[25]  Shahaboddin Shamshirband,et al.  Comparative analysis of reference evapotranspiration equations modelling by extreme learning machine , 2016, Comput. Electron. Agric..

[26]  Witold R. Rudnicki,et al.  Boruta - A System for Feature Selection , 2010, Fundam. Informaticae.

[27]  Wieslaw Paja,et al.  All Relevant Feature Selection Methods and Applications , 2015, Feature Selection for Data and Pattern Recognition.

[28]  A.J.M. Timmermans,et al.  Computer vision system for on-line sorting of pot plants using an artificial neural network classifier , 1996 .

[29]  O. Carisse,et al.  A review of non-fungicidal approaches for the control of apple scab , 2002 .

[30]  Shervin Motamedi,et al.  Extreme learning machine based prediction of daily dew point temperature , 2015, Comput. Electron. Agric..

[31]  F. Ronsse,et al.  Estimation of leaf wetness duration for greenhouse roses using a dynamic greenhouse climate model in Zimbabwe , 2013 .

[32]  Jun Wang,et al.  Discrimination among tea plants either with different invasive severities or different invasive times using MOS electronic nose combined with a new feature extraction method , 2017, Comput. Electron. Agric..

[33]  Hanno Scharr,et al.  A cognitive architecture for automatic gardening , 2017, Comput. Electron. Agric..

[34]  B. Meszka,et al.  The status of dodine resistance of Venturia inaequalis populations in Poland. , 2008 .

[35]  Krishna N. Reddy,et al.  Random forest and leaf multispectral reflectance data to differentiate three soybean varieties from two pigweeds , 2016, Comput. Electron. Agric..

[36]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[37]  Yoav Freund,et al.  A decision-theoretic generalization of on-line learning and an application to boosting , 1997, EuroCOLT.

[38]  W. E. Machardy,et al.  Apple Scab: Biology, Epidemiology, and Management , 1996 .

[39]  Ramón Díaz-Uriarte,et al.  Gene selection and classification of microarray data using random forest , 2006, BMC Bioinformatics.

[40]  A. Gayen.  Significance of difference between the means of two non-normal samples. , 1950, Biometrika.

[41]  Dayang Liu,et al.  Peach variety identification using near-infrared diffuse reflectance spectroscopy , 2016, Comput. Electron. Agric..

[42]  W. Schwabe.  Wetting and temperature requirements for infection of mature apples by Venturia inaequalis in South Africa , 1982 .

[43]  Jason H. Moore,et al.  Learning classifier systems: a complete introduction, review, and roadmap , 2009 .

[44]  Zixiang Xiong,et al.  Optimal number of features as a function of sample size for various classification rules , 2005, Bioinform..

[45]  T. Davies,et al.  A laboratory assessment of wetness sensors for leaf, fruit and trunk surfaces. , 2000 .

[46]  J. Weres,et al.  Neural identification of selected apple pests , 2015, Comput. Electron. Agric..

[47]  Z. Parveen,et al.  Evaluation of Multiple Pesticide Residues in Apple and Citrus Fruits, 1999–2001 , 2004, Bulletin of environmental contamination and toxicology.

[48]  Welch Bl.  THE GENERALIZATION OF ‘STUDENT'S’ PROBLEM WHEN SEVERAL DIFFERENT POPULATION VARLANCES ARE INVOLVED , 1947 .

[49]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[50]  D. Gadoury,et al.  A revision of Mill's criteria for predicting apple scab infection periods. , 1989 .