Optimalno navodnjavanje pšenice primjenom umjetne inteligencije

master's thesis

• Title (english): Optimal wheat irrigation by artificial intelligence
• Author: Edi Slijepčević
• Mentor: Vinko Lešić (mentor)
• Committee member: Vinko Lešić (član povjerenstva)
• Committee member: Mato Baotić (član povjerenstva)
• Committee member: Branimir Novoselnik (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-09, Croatia

Abstract

Kako bi bila osigurana veća sigurnost opskrbe hranom potrebno je razvijati nove metode upravljanja i kontrole rasta i razvoja usjeva s naglaskom na navodnjavanje. U ovu svrhu razvijen je model povratne neuronske mreže koji predviđa vrijednosti NDVI-a i površine pšenice koja je korištena kao alternativa za biomasu. Model je testiran na stvarnim podacima s terena te se pokazao kao zadovoljavajuć s obzirom na opseg dostupnih podataka. Istražen je i remote-sensing pristup koji obuhvaća prikupljanje podataka o NDVI vrijednostima korištenjem Sentinel 2 satelita te podataka o prognozi korištenjem Open-Weather platforme te uparivanje tih vrijednosti sa stvarnim mjerenjima na terenu kako bi se dobila što bolja predikcija ciljanih parametara.

Abstract (english)

In order to ensure greater security of food supply, it is necessary to develop new methods of managing and controlling the growth and development of crops with an emphasis on irrigation. For this purpose, a recurrent neural network model was developed that predicts the NDVI values and the area of wheat plant that was used as an alternative for biomass. The model was tested on real data from the field and proved to be satisfactory considering the extent of available data. A remote-sensing approach was also investigated, which includes collecting data on NDVI values using the Sentinel 2 satellite and forecast data using the Open-Weather platform, and matching these values with actual measurements in the field in order to obtain the best possible prediction of target parameters.

• Keywords: pšenica; normalizirani razlikovni indeks; biomasa; povratna neuronska mreža; navodnjavanje; vremenska prognoza; remote-sensing
• Keywords (english): wheat; normalized difference vegetation index; biomass; recurrent neural network; irrigation; weather; remote-sensing
• Language: croatian
• URN:NBN: urn:nbn:hr:168:934337
• Study programme: Title: Information and Communication Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije (magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2024-10-12 22:07:13