Mladen Ivanišević

Dr. Mladen Ivanišević

orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

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

Representative Publications

View All Publications
16 July 2025 in Agriculture
Verified
Journal Article
Open access
PDF file
Krstan KešeljKrstan Kešelj
Krstan Kešelj
ZSZoran Stamenković
Zoran Stamenković
Marko Milan KostićMarko Milan Kostić
Marko Milan Kostić
Vladimir AćinVladimir Aćin
Vladimir Aćin
Dragana NovakovićDragana Novaković
Dragana Novaković
NMNenad Magazin
Nenad Magazin
Accurate and timely wheat yield prediction is valuable globally for enhancing agricultural planning, optimizing resource use, and supporting trade strategies. Study addresses the need for precision in yield estimation by applying machine-learning (ML) regression models to high-resolution Unmanned Aerial Vehicle (UAV) multispectral (MS) and Red-Green-Blue (RGB) imagery. Research analyzes five European wheat cultivars across 400 experimental plots created by combining 20 nitrogen, phosphorus, and potassium (NPK) fertilizer treatments. Yield variations from 1.41 to 6.42 t/ha strengthen model robustness with diverse data. The ML approach is automated using PyCaret, which optimized and evaluated 25 regression models based on 65 vegetation indices and yield data, resulting in 66 feature variables across 400 observations. The dataset, split into training (70%) and testing sets (30%), was used to predict yields at three growth stages: 9 May, 20 May, and 6 June 2022. Key models achieved high accuracy, with the Support Vector Regression (SVR) model reaching R2 = 0.95 on 9 May and R2 = 0.91 on 6 June, and the Multi-Layer Perceptron (MLP) Regressor attaining R2 = 0.94 on 20 May. The findings underscore the effectiveness of precisely measured MS indices and a rigorous experimental approach in achieving high-accuracy yield predictions. This study demonstrates how a precise experimental setup, large-scale field data, and AutoML can harness UAV and machine learning’s potential to enhance wheat yield predictions. The main limitations of this study lie in its focus on experimental fields under specific conditions; future research could explore adaptability to diverse environments and wheat varieties for broader applicability.

1 December 2024 in Contemporary Agriculture
Verified
Journal Article
Open access
PDF file
ZSZoran Stamenković
Zoran Stamenković
Krstan KešeljKrstan Kešelj
Krstan Kešelj
Marko Milan KostićMarko Milan Kostić
Marko Milan Kostić
Vladimir AćinVladimir Aćin
Vladimir Aćin
Dragana NovakovićDragana Novaković
Dragana Novaković
TNTihomir Novaković
Tihomir Novaković
Summary: Precision agriculture has increasingly incorporated the use of Unmanned Aerial Vehicles (UAVs) equipped with multispectral cameras. This study examined the influence of different UAV flight altitudes on the Normalized Difference Vegetation Index (NDVI), Green Normalized Difference Vegetation Index (GNDVI), Normalized Difference Red Edge (NDRE), Optimized Soil-Adjusted Vegetation Index (OSAVI), and Leaf Chlorophyll Index (LCI), indices critical to crop monitoring and health assessment. The experiment was conducted on a 2-hectare winter wheat field at the Institute of Field and Vegetable Crops in Novi Sad, Serbia. The field was divided into 400 plots, each containing different wheat varieties subjected to twenty distinct combinations of artificial mineral fertilizer (NPK) treatments. A DJI P4 Multispectral drone was employed to capture images at altitudes of 30, 60, and 90 meters on three separate dates, corresponding to different plant growth stages: May 9, May 20, and June 6, 2022. All other operating parameters were held constant. The data were processed using the DJI Terra and Pix4D software to generate orthomosaic maps, which were subsequently analyzed using ArcGIS (v10.5, ESRI, Redlands, CA, USA) to calculate the multispectral index values for each plot. The results were statistically analyzed using the STATISTICA Tibco software. The analysis revealed significant differences in the index values based on the UAV flight altitude (p < 0.05). This research underscores the centrality of selecting the optimal UAV flight altitude to ensure the accuracy and reliability of data. While higher altitudes enable UAVs to cover larger areas in a single flight, factors such as image resolution, wind conditions, and the precision of crop health indicators must be considered. These findings offer valuable insights for agricultural professionals seeking to improve crop monitoring and ultimately enhance agricultural productivity through more effective UAV deployment.

16 February 2026 in Agriengineering
Verified
Journal Article
Open access
PDF file
Krstan KešeljKrstan Kešelj
Krstan Kešelj
ZSZoran Stamenković
Zoran Stamenković
Marko Milan KostićMarko Milan Kostić
Marko Milan Kostić
Vladimir AćinVladimir Aćin
Vladimir Aćin
AIAleksandar Ivezić
Aleksandar Ivezić
NMNenad Magazin
Nenad Magazin
Accurate wheat yield prediction is essential for ensuring food security and sustainable resource management under the increasing challenges of climate change. This study investigates the integration of unmanned aerial vehicle (UAV)-based multispectral imaging and machine learning (ML) techniques to improve yield forecasting in European wheat cultivars. Field experiments were conducted on 400 sub-plots with varying NPK fertilization regimes and five wheat varieties, monitored across six phenological stages during the 2023 growing season in Vojvodina, Serbia. A DJI Phantom 4 Multispectral UAV collected high-resolution imagery, from which 65 vegetation indices were computed. Using PyCaret’s automated ML framework, 25 regression algorithms were evaluated for yield prediction. Ensemble models, particularly Random Forest, Extra Trees, Gradient Boosting, and LightGBM, consistently outperformed linear and kernel-based approaches. The highest prediction accuracy was achieved with the Random Forest Regressor during full flowering (BBCH 65–69), yielding an R2 of 0.952 and an RMSE of 0.44 t/ha. Results highlight the temporal dynamics of model performance, with optimal predictions occurring during reproductive stages. The findings confirm that UAV-derived multispectral data, coupled with ensemble machine learning, provide a non-invasive, accurate, and computationally efficient method for yield forecasting. This framework has significant potential for supporting precision agriculture, enabling real-time decision-making, and enhancing the resilience of wheat production systems.

1 January 2022 in Thermal Science
Verified
Journal Article
Open access
PDF file
MIMladen Ivanišević
Mladen Ivanišević
MZMiodrag S. Zoranović
Miodrag S. Zoranović
GTGoran R. Topisirović
Goran R. Topisirović
MJMilan A. Jugović
Milan A. Jugović
VRVladimir Rajs
Vladimir Rajs
Krstan KešeljKrstan Kešelj
Krstan Kešelj
The object of this research is the new concept of the original universal system for air quality and energy control in laboratory conditions (wet scrubbing system) based on the ?clean air in/out? principle. Namely, the process involves a partial or complete exchange of recirculated treated air. From the aspect of water, as a pure medium, the system is characterized by broad application in the open, semi-open, and closed air treatment systems: primarily in the breeding and accompanying agricultural facilities, industrial, cultural, sports, tourist, medical and other controlled spaces. The analysis showed that the system at its inlet unit and with the block structure and logical phase development, proved energy, environ-mental and qualitative efficiency in the removal of PM10 and PM2.5 particles from the controlled dust doses in laboratory air. The measurements of PM10 and PM2.5 concentrations were performed with two Alphasense OPC-N3 optical counters at the inlet and outlet of the system. Five operating regimes with frequency regulated number of revolutions of the turbo elements that was taken as an independent variable, achieved different degrees of removal of PM10 and PM2.5 from the treated laboratory air: 97.70-98.53% and 62.19-75.75%, respectively. The assessment of system energy use was done by parallel measuring of electric power and comparative deviations of its absolute values in the first decimal. Energy consumption for the treatment of 1m3 of air ranged from 0.00011- 0.00016 kWm?3. Statistical analysis of qualitative indicators revealed significant differences between the operating regimes and the obtained values.

9 June 2023 in Notulae Botanicae Horti Agrobotanici Cluj-Napoca
Verified
Journal Article
Open access
PDF file
MJMilan Jugovic
Milan Jugovic
MZMiodrag Zoranovic
Miodrag Zoranovic
MIMladen Ivanišević
Mladen Ivanišević
ZMZoran Malicevic
Zoran Malicevic
MZMilovan Zivkovic
Milovan Zivkovic
TJTanja Jakisic
Tanja Jakisic
The paper presents the results of three different ways of storing the dried mass of hay: bulk, small bales and large roll bales, as well as the impact of three drying methods: natural in the field, artificial drying with cold air and drying with dehumidification (warm air). In the tested meadow in the first swath, the results of chemical analyses showed differences in the method of drying hay. Regarding the tested drying method, the content of dry matter (DM) had significant differences between the storage methods as well as all variants with pre-heated air drying, where the average value of DM was in the interval of 86.18-93.01%. The content of mineral substances for certain methods of preparation and drying ranged from 5.77% to 7.72% on average. The highest content of crude proteins was in all variants of artificial drying and it ranged from 98.6 to 165.7 g/kg DM and had a statistically significant difference. Both methods of artificial post-drying had a significant impact on the cellulose content (33.76% to 28.86%) compared to drying in the traditional way because postponing the mowing time increases the cellulose content. The drying method had a statistically high significant difference on the content of neutral detergent fibres (NDF) and acidic detergent fibres (ADF), while the method of storage had no major impact. Knowledge of changes in the quality of hay during the growing season is of particular importance form the aspect of ruminant nutrition and balanced rations. The amount and quality of obtained hay is significantly affected by the time of mowing, height of mowing, swath, fertilization, floristic composition and weather conditions during drying of the green mass.

1 January 2016 in Savremena poljoprivredna tehnika
Verified
Journal Article
Open access
PDF file
MIMladen Ivanišević
Mladen Ivanišević
SMSnežana Matić-Kekić
Snežana Matić-Kekić
NDN. Dedović
N. Dedović
MZM. Zoranović
M. Zoranović
VVV. Višacki
V. Višacki
For shape analysis of plants and their parts is used principal components analysis of elliptical Furieovih descriptors. As a result of the analysis are obtained independent features of shape - the principal component which enabled quantified analysis of shape. For the shape analysis it was randomly selected fifty varieties of pear fruits Williams. The fruits are placed on a dark background in a reclined position. Dark background creates a good contrast between the background and fruit, making it easier to detect the contours of pears. Variations shape of a pear can be covered with the principal component 1 and 2 (PC1 and PC2). It also shows that the principal components 3, 4 and 5 (PC3, PC4 and PC5) have very little shape variation so that they can be ignored.