Seite - 130 - in Intelligent Environments 2019 - Workshop Proceedings of the 15th International Conference on Intelligent Environments

AnLSTMDeepLearningSchemefor M.A´ngelGUILLE´N-NAVARROa,Raquel MARTI´NEZ-ESPAN˜Aa, Andre´sBUENO-CRESPOaBele´nAYUSOa JoseLuis MORENOa and Jose´M.CECILIAa aDept. ofComputerEngineering,CatholicUniversityofMurcia,Murcia,Spain Abstract. Precision agriculture adopts a set of techniques capable of increasing productivity,yieldandefficiencyinworkrelated toagriculture,producingagreater benefit for farmers. In this studywe focus on the problem of predictingweather conditions, specifically thepredictionof lowtemperatures.Temperatureprediction isamajorprobleminagriculture.Farmerscan lose their crops if frost control tech- niques are not activated in time. The threshold for activating such techniques de- pendsonthe typeofcrop.Afirstpreliminarystudyusingdeeplearningisproposed topredict temperature, particularly aLongShort-TermMemoryNetwork (LSTM) isused.TheLSTMhasbeen trainedusingreal temperaturedataprovidedbyan the Internet of Things (IoT) system, deployed in several plots and currently in opera- tion. The results obtained after testing themodel createdwith this neural network arequite satisfactoryobtainingadeterminationcoefficient (R2)of99%andanav- erage quadratic error of less than 0.8 degrees Celsius. Given the goodness of the model thiscanbe implementedasan intelligentcomponentof the IoTsystem, thus complementing its functionality. Keywords. temperatureprediction,precisionagriculture, deep learning, regresion, LSTM 1. Introduction The frameworkofPrecisionAgriculture canbedefined as a systemic approach to reor- ganize theentireagriculturalsystemtowardssustainable,high-efficiency, low-inputagri- culture[14,21].Thisnewapproachbenefitsmainlyfromtheemergenceandconvergence of new technologies, including theGlobal PositioningSystem, thegeographic informa- tionsystem,sensors,automaticcontrol, remotesensingandremotesensing,mobilecom- puting, advanced information processing and telecommunications. All these technolo- gies unified under the paradigmofCloudComputing and the Internet of Things (IoT) allowus to efficiently and easily deploy new technologieswithin the various areas that affect agriculture [18].Therefore,precisionagriculturealsoprovidesabenefit to society sinceanytechnique that facilitates thestabilizationor increaseofagriculturalproduction and at the same timemediate in the environmental impacts of this activity [20]. Some of the activities involvedwithin precision agriculture are the disease detection [12,15], PredictionofLowTemperatures in Agriculture Intelligent Environments 2019 A. Muñoz et al. (Eds.) © 2019 The authors and IOS Press. This article is published online with Open Access by IOS Press and distributed under the terms of the Creative Commons Attribution Non-Commercial License 4.0 (CC BY-NC 4.0). doi:10.3233/AISE190032 130