Rajko Miodragović

Rajko Miodragović

Auto-generated profile, based on publicly available information

Rajko Miodragović

University of Belgrade

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 January 2024 in Poljoprivredna tehnika
Verified
Journal Article
Open access
PDF file
OEOlivera Ecim-Djuric
Olivera Ecim-Djuric
RMRajko Miodragović
Rajko Miodragović
ARAndrija Rajković
Andrija Rajković
MMMihailo Milanović
Mihailo Milanović
ZMZoran Mileusnić
Zoran Mileusnić
ADAleksandra Ljubodrag Dragicevic
Aleksandra Ljubodrag Dragicevic
Machine learning (ML) is a key technology driving the modernisation of agriculture. It enables large data sets to be analysed and precise decisions to be made at all stages of agricultural production. ML is used for soil analysis, plant disease detection, weed control, crop species identification and harvest optimisation. Various techniques such as supervised, unsupervised and reinforcement learning help to improve the accuracy of predictions and decisions. Artificial neural networks (ANN), in particular deep neural networks (DNN) and convolutional neural networks (CNN), efficiently analyse images and numerical data and enable precise management of agricultural practises. These technologies contribute to sustainability by reducing the negative impact on the environment and optimising the use of resources. While significant progress has already been made, there is still potential for further development of ML models that cover all phases of the agricultural cycle and make precision agriculture more efficient and safer.

1 January 2024
Verified
Book Chapter
Closed access
PDF file
OEOlivera Ecim-Djuric
Olivera Ecim-Djuric
MMMihailo Milanovic
Mihailo Milanovic
ADAleksandra Ljubodrag Dragicevic
Aleksandra Ljubodrag Dragicevic
ARAndrija Rajković
Andrija Rajković
ZMZoran Mileusnić
Zoran Mileusnić
RMRajko Miodragović
Rajko Miodragović

1 January 2025
Verified
Conference Paper
Open access
PDF file
Vjekoslav TadićVjekoslav Tadić
Vjekoslav Tadić
ASA. Stanisavljević
A. Stanisavljević
TKT. Kujundžić
T. Kujundžić
DBD. Bošnjak
D. Bošnjak
ASA. Strilić
A. Strilić
RMR. Miodragović
R. Miodragović
The project proposal envisages a four-year study (2 years' apple orchard and 2 years' vineyard) to investigate the influence of technical and technological factors of pesticide application on the main characteristics of the study: Residues of active pesticide substances in the fruit, occurrence and spread of pests, fruit quality, and deposition and drift of the spray. The technical and technological factors of the application are the spraying rate, the spray concentration and the nozzle type with its sub-factors. The spraying rate is determined according to the TRV method (tree row volume) and a 50% reduction, while the concentration is determined according to the recommended (FIS registration) and its 20 and 40% reduction. According to the technological map for the protection of permanent plantations, two types of nozzles are used for spraying: Standard TR and Air-Injector ITR. The residues of the active substances are determined using the liquid chromatography with mass spectrometer (LC-MS/MS) method in an accredited laboratory, while the quality of the fruit and the occurrence and development of the pests are determined in the Faculty of agrobiotechnical sciences central laboratory. The deposit and spray drift are determined using the spectrophotometric method defined by the ISO 22866:2005 standard. The field trial should provide information on whether it is possible to reduce the application rate and concentration of sprays (low-volume pesticide management) and at the same time improve the quality of the fruit with the same or better biological effect on the pests. In addition, the improved quality of spraying will reduce drift, which will have an impact on better pest control and the possibility of reducing the number of applications.