Klasifikacija i analiza poljoprivrednih zemljišta u Donjem Međimurju pomoću podataka misija Sentinel

master's thesis

• Title (english): Agricultural areas classification and analysis in the Donje Međimurje region using Sentinel data
• Author: Josipa Golomboš
• Mentor: Luka Valožić (mentor)
• Committee member: Luka Valožić (predsjednik povjerenstva)
• Committee member: Mladen Pahernik (član povjerenstva)
• Committee member: Dubravka Spevec (član povjerenstva)
• Granter: University of Zagreb; Faculty of Science; (Department of Geography); Zagreb
• Defense date and country: 2022-02-24, Croatia
• Scientific / art field, discipline and subdiscipline: INTERDISCIPLINARY AREAS OF KNOWLEDGE; Geography

Abstract

Cilj ovog istraživanja je random forest klasifikacijom izdvojiti poljoprivredna zemljišta i klasificirati ih prema 6 klasa poljoprivrednih usjeva. Korištene su optičke snimke visoke rezolucije Sentinel 2 satelitske misije, a preuzeto je po 5 snimaka za 2017., 2018., 2019. i 2020. godinu. Sentinel-2 je satelitska misija koja se sastoji od satelita S-2A i S-2B koja snima Zemljinu površinu i formira satelitske snimke sa 13 spektralnih kanala prostornih rezolucija od 10 m, 20 m i 60 m. Snimke su prethodno obrađene i klasificirane pomoću
programa SNAP, a dodatna vizualizacija kao i matrica pogrške je izrađena u ArcGIS Pro programu. U ovom radu će se analizirati promjene u odabiru usjeva pojedine godine, raspravljati će se o godišnjim cijenama pojedinih usjeva te utjecaju određenih faktora na odabir usjeva.

Abstract (english)

The goal of this research is to extract agricultural lands and classify them according to 6 types of agricultural crops. High-resolution optical images of the Sentinel 2 satellite mission were used and 5 images were taken for the years 2017, 2018, 2019 and 2020. Sentinel-2 is a satellite mission consisting of satellites S-2A and S-2B that captures the Earth's surface and forms satellite images with 13 spectral channels of spatial resolutions of 10 m, 20 m and 60 m. The images are pre-processed and classified using SNAP software Additional visualization as well as an accuracy matrix was created in the ArcGIS Pro
software. This paper will analyze changes in crop selection in a particular year, discuss the annual prices of individual crops and the impact of different on crop selection.

• Keywords: SNAP; random forest; poljoprivredna zemljišta; Sentinel-2; Donje Međimurje
• Keywords (english): SNAP; random forest; agricultural area; Sentinel-2; Donje Međimurje
• Language: croatian
• URN:NBN: urn:nbn:hr:217:146853
• Study programme: Title: Geography; specializations in: Physical Geography with Geoecology, Spatial Planning and Regional Development, Heritage and Tourism, Geographic Information Systems, Course research; Course: Geographic Information Systems; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra geografije (magistar/magistra geografije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2022-03-08 11:42:20