Primjena računalnog vida za pomoć u usmjerenoj aplikaciji herbicida

master's thesis

• Title (english): The use of computer vision in variable rate herbicide application
• Author: Sinan Adakaya
• Mentor: Igor Kovačev (mentor)
• Committee member: Igor Kovačev (predsjednik povjerenstva)
• Committee member: Ante Galić (član povjerenstva)
• Committee member: Maja Šćepanović (član povjerenstva)
• Committee member: Zlatko Koronc (član povjerenstva)
• Granter: University of Zagreb; Faculty of Agriculture; (Mechanization and Autonomous Systems in Agriculture); Zagreb
• Defense date and country: 2018-06-14, Croatia
• Scientific / art field, discipline and subdiscipline: BIOTECHNICAL SCIENCES; Agronomy

Abstract

Ovaj je rad nastojao prikazati nove tehnologije razvijene za kontrolu korova. Korov predstavlja veliki problem u ostvarivanju maksimalnih prinosa, što je u današnjoj poljoprivredi od iznimne važnosti. Posebice u fokusu je aktualno stanje računalnog vida i obrade slike za pomoć u usmjerenoj aplikaciji herbicida. Istaknuto je stanje kontrole korova u svijetu, pozitivni i negativni aspekti aktualnih metoda te razlozi zbog kojih su te metode još uvijek u upotrebi. Nadalje, pružajući pregled važnih ranijih istraživanja, prikazana su mišljenja istraživača, korišteni materijali i metode te rezultati njihovih istraživanja. Prikazani su najčešći izazovi na koje istraživači nailaze pri korištenju novih metoda te smjer mogućih budućih istraživanja u području računalne obrade slika i robotske kontrole korova. Cilj rada bio je stvoriti pregled koji će služiti kao korisna kolekcija recentne literature istraživačima za buduća istraživanja.

Abstract (english)

The aim of this review was to bring forth new technologies developed for weed control. Weed poses a big risk to maximizing yield, which is of utmost importance in today's agriculture. In particular, we focused on the current state of computer vision and image processing in assisting variable rate herbicide application. The state of weed control in the world, positive and negative aspects of current methods and the reasons why these methods are still in use are pointed out in this review. Furthermore, by providing an overview of important previous studies, the view of researchers, used materials and methods, as well as results of their studies were laid out. The most common challenges that researchers encounter when using new methods were discussed. The direction of possible future studies in the field of image processing and robotic weed control was proposed. The overall aim of the paper was to create a review that will serve as a useful collection of recent literature to help researchers with future research.

• Keywords: kontrola korova; računalna obrada slike; robotska kontrola korova; usmjerena aplikacija herbicida
• Keywords (english): weed control; image processing; robotic weed control; variable rate application
• Language: croatian
• URN:NBN: urn:nbn:hr:204:195085
• Study programme: Title: Agricultural Engineering; Study programme type: university; Study level: graduate; Academic / professional title: magistar/ magistra inženjer/inženjerka poljoprivredne tehnike (magistar/ magistra inženjer/inženjerka poljoprivredne tehnike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Access restricted to students and staff of home institution
• Created on: 2018-06-21 06:20:59