TY  - JOUR
AU  - Potić, Ivan
AU  - Srdić, Zoran
AU  - Vakanjac, Boris
AU  - Bakrač, Saša
AU  - Đorđević, Dejan
AU  - Banković, Radoje
AU  - Jovanović, Jasmina M.
PY  - 2023
UR  - http://gery.gef.bg.ac.rs/handle/123456789/1719
AB  - Vegetation plays an active role in ecosystem dynamics, and monitoring its patterns and changes is vital for effective environmental resource management. This study explores the possibility of machine learning techniques and remote sensing data to improve the accuracy of forest detection. The research focuses on the southeastern part of the Republic of Serbia as a case study area, using Sentinel-2 multispectral bands. The study employs publicly accessible satellite data and incorporates different vegetation indices to improve classification accuracy. The main objective is to examine the practicability of expanding the input parameters for forest detection using a machine learning approach. The classification process is performed by employing support vector machines (SVM) algorithm and utilising the SVM module in the scikit-learn package. The results demonstrate that including vegetation indices alongside the multispectral bands significantly improves the accuracy of vegetation detection. A comprehensive assessment reveals an overall classification accuracy of up to 99.01% when the selected vegetation indices (MCARI, RENDVI, NDI45, GNDVI, NDII) are combined with the Sentinel-2 bands. This research highlights the potential of machine learning and remote sensing in forest detection and monitoring. The findings underscore the importance of incorporating vegetation indices to enhance classification accuracy using the Python programming language. The study’s outcomes provide valuable insights for environmental resource management and decision-making processes, particularly in regions with diverse forest ecosystems.
PB  - Basel : MDPI
T2  - Applied Sciences
T1  - Improving Forest Detection Using Machine Learning and Remote Sensing: A Case Study in Southeastern Serbia
VL  - 13
IS  - 14
SP  - 8289
DO  - 10.3390/app13148289
ER  - 

@article{
author = "Potić, Ivan and Srdić, Zoran and Vakanjac, Boris and Bakrač, Saša and Đorđević, Dejan and Banković, Radoje and Jovanović, Jasmina M.",
year = "2023",
abstract = "Vegetation plays an active role in ecosystem dynamics, and monitoring its patterns and changes is vital for effective environmental resource management. This study explores the possibility of machine learning techniques and remote sensing data to improve the accuracy of forest detection. The research focuses on the southeastern part of the Republic of Serbia as a case study area, using Sentinel-2 multispectral bands. The study employs publicly accessible satellite data and incorporates different vegetation indices to improve classification accuracy. The main objective is to examine the practicability of expanding the input parameters for forest detection using a machine learning approach. The classification process is performed by employing support vector machines (SVM) algorithm and utilising the SVM module in the scikit-learn package. The results demonstrate that including vegetation indices alongside the multispectral bands significantly improves the accuracy of vegetation detection. A comprehensive assessment reveals an overall classification accuracy of up to 99.01% when the selected vegetation indices (MCARI, RENDVI, NDI45, GNDVI, NDII) are combined with the Sentinel-2 bands. This research highlights the potential of machine learning and remote sensing in forest detection and monitoring. The findings underscore the importance of incorporating vegetation indices to enhance classification accuracy using the Python programming language. The study’s outcomes provide valuable insights for environmental resource management and decision-making processes, particularly in regions with diverse forest ecosystems.",
publisher = "Basel : MDPI",
journal = "Applied Sciences",
title = "Improving Forest Detection Using Machine Learning and Remote Sensing: A Case Study in Southeastern Serbia",
volume = "13",
number = "14",
pages = "8289",
doi = "10.3390/app13148289"
}

Potić, I., Srdić, Z., Vakanjac, B., Bakrač, S., Đorđević, D., Banković, R.,& Jovanović, J. M.. (2023). Improving Forest Detection Using Machine Learning and Remote Sensing: A Case Study in Southeastern Serbia. in Applied Sciences
Basel : MDPI., 13(14), 8289.
https://doi.org/10.3390/app13148289

Potić I, Srdić Z, Vakanjac B, Bakrač S, Đorđević D, Banković R, Jovanović JM. Improving Forest Detection Using Machine Learning and Remote Sensing: A Case Study in Southeastern Serbia. in Applied Sciences. 2023;13(14):8289.
doi:10.3390/app13148289 .

Potić, Ivan, Srdić, Zoran, Vakanjac, Boris, Bakrač, Saša, Đorđević, Dejan, Banković, Radoje, Jovanović, Jasmina M., "Improving Forest Detection Using Machine Learning and Remote Sensing: A Case Study in Southeastern Serbia" in Applied Sciences, 13, no. 14 (2023):8289,
https://doi.org/10.3390/app13148289 . .

TY  - JOUR
AU  - Jovanović, Jasmina M.
AU  - Stojanović, Marko
AU  - Janković, Tanja
AU  - Drobnjak, Siniša
AU  - Djordjević, Dejan
AU  - Banković, Radoje
AU  - Radovanović, Milan
AU  - Živković, Ljiljana
AU  - Gajić, Tamara
AU  - Demirović Bajrami, Dunja
AU  - Tretiakova, Tatiana N.
AU  - Syromiatnikova, Julia A.
PY  - 2022
UR  - http://gery.gef.bg.ac.rs/handle/123456789/1413
AB  - Cultural assets in the area of the Danube Limes in Serbia are an integral part of the world
heritage “Roman Empire Borders”. The research presented in this paper includes the tourist and
cartographic visualization of 19 Roman sites in the Danube Limes region of Golubac–Radujevac,
to determine the real possibilities of tourism development in this area. The historical and cultural
heritage of this area is among the most attractive tourist destinations in Serbia, Djerdap National
Park and Djerdap Geopark. Despite its diverse cultural and historical values and the specific and
unique natural environment, this area is not sufficiently used for tourism. The research included
the evaluation of localities, which may serve as the basis to establish which activities should be
undertaken in order to plan, use, preserve, and protect such important cultural assets, under the
principles of sustainable tourism development. Information based on spatially referenced data in the
research process requires cartographic support, in order to understand the geospatial relations of
the site significance. Cartographic visualization enabled efficiently systematized data organization,
spatial identification, presentation, and the use of complex information from the mapped area in the
data analysis in this paper.
PB  - Basel : MDPI
T2  - Sustainability
T1  - A Case Study on the Danube Limes in Serbia: Valorisation and Cartographic Analyses of Selected Tourism Products
VL  - 14
IS  - 3
SP  - 1480
DO  - 10.3390/su14031480
ER  - 

@article{
author = "Jovanović, Jasmina M. and Stojanović, Marko and Janković, Tanja and Drobnjak, Siniša and Djordjević, Dejan and Banković, Radoje and Radovanović, Milan and Živković, Ljiljana and Gajić, Tamara and Demirović Bajrami, Dunja and Tretiakova, Tatiana N. and Syromiatnikova, Julia A.",
year = "2022",
abstract = "Cultural assets in the area of the Danube Limes in Serbia are an integral part of the world
heritage “Roman Empire Borders”. The research presented in this paper includes the tourist and
cartographic visualization of 19 Roman sites in the Danube Limes region of Golubac–Radujevac,
to determine the real possibilities of tourism development in this area. The historical and cultural
heritage of this area is among the most attractive tourist destinations in Serbia, Djerdap National
Park and Djerdap Geopark. Despite its diverse cultural and historical values and the specific and
unique natural environment, this area is not sufficiently used for tourism. The research included
the evaluation of localities, which may serve as the basis to establish which activities should be
undertaken in order to plan, use, preserve, and protect such important cultural assets, under the
principles of sustainable tourism development. Information based on spatially referenced data in the
research process requires cartographic support, in order to understand the geospatial relations of
the site significance. Cartographic visualization enabled efficiently systematized data organization,
spatial identification, presentation, and the use of complex information from the mapped area in the
data analysis in this paper.",
publisher = "Basel : MDPI",
journal = "Sustainability",
title = "A Case Study on the Danube Limes in Serbia: Valorisation and Cartographic Analyses of Selected Tourism Products",
volume = "14",
number = "3",
pages = "1480",
doi = "10.3390/su14031480"
}

Jovanović, J. M., Stojanović, M., Janković, T., Drobnjak, S., Djordjević, D., Banković, R., Radovanović, M., Živković, L., Gajić, T., Demirović Bajrami, D., Tretiakova, T. N.,& Syromiatnikova, J. A.. (2022). A Case Study on the Danube Limes in Serbia: Valorisation and Cartographic Analyses of Selected Tourism Products. in Sustainability
Basel : MDPI., 14(3), 1480.
https://doi.org/10.3390/su14031480

Jovanović JM, Stojanović M, Janković T, Drobnjak S, Djordjević D, Banković R, Radovanović M, Živković L, Gajić T, Demirović Bajrami D, Tretiakova TN, Syromiatnikova JA. A Case Study on the Danube Limes in Serbia: Valorisation and Cartographic Analyses of Selected Tourism Products. in Sustainability. 2022;14(3):1480.
doi:10.3390/su14031480 .

Jovanović, Jasmina M., Stojanović, Marko, Janković, Tanja, Drobnjak, Siniša, Djordjević, Dejan, Banković, Radoje, Radovanović, Milan, Živković, Ljiljana, Gajić, Tamara, Demirović Bajrami, Dunja, Tretiakova, Tatiana N., Syromiatnikova, Julia A., "A Case Study on the Danube Limes in Serbia: Valorisation and Cartographic Analyses of Selected Tourism Products" in Sustainability, 14, no. 3 (2022):1480,
https://doi.org/10.3390/su14031480 . .

TY  - JOUR
AU  - Drobnjak, Siniša
AU  - Stojanović, Marko
AU  - Djordjević, Dejan
AU  - Bakrač, Saša
AU  - Jovanović, Jasmina
AU  - Djordjević, Aleksandar
PY  - 2022
UR  - http://gery.gef.bg.ac.rs/handle/123456789/1415
AB  - The objective of this research is to report results from a new ensemble method for
vegetation classification that uses deep learning (DL) and machine learning (ML)
techniques. Deep learning and machine learning architectures have recently been used
in methods for vegetation classification, proving their efficacy in several scientific
investigations. However, some limitations have been highlighted in the literature, such
as insufficient model variance and restricted generalization capabilities. Ensemble DL and
ML models has often been recommended as a feasible method to overcome these
constraints. A considerable increase in classification accuracy for vegetation classification
was achieved by growing an ensemble of decision trees and allowing them to vote for the
most popular class. An ensemble DL and ML architecture is presented in this study to
increase the prediction capability of individual DL and ML models. Three DL and ML
models, namely Convolutional Neural Network (CNN), Random Forest (RF), and biased
Support vector machine (B-SVM), are used to classify vegetation in the Eastern part of
Serbia, together with their ensemble form (CNN-RF-BSVM). The suggested DL and ML
ensemble architecture achieved the best modeling results with overall accuracy values
(0.93), followed by CNN (0.90), RF (0.91), and B-SVM (0.88). The results showed that the
suggested ensemble model outperformed the DL and ML models in terms of overall
accuracy by up to 5%, which was validated by the Wilcoxon signed-rank test. According to
this research, RF classifiers require fewer and easier-to-define user-defined parameters
than B-SVMs and CNN methods. According to overall accuracy analysis, the proposed
ensemble technique CNN-RF-BSVM also significantly improved classification
accuracy (by 4%).
PB  - Frontiers Media SA
T2  - Frontiers in Environmental Science
T1  - Testing a New Ensemble Vegetation Classification Method Based on Deep Learning and Machine Learning Methods Using Aerial Photogrammetric Images
VL  - 10
SP  - 896158
DO  - 10.3389/fenvs.2022.896158
ER  - 

@article{
author = "Drobnjak, Siniša and Stojanović, Marko and Djordjević, Dejan and Bakrač, Saša and Jovanović, Jasmina and Djordjević, Aleksandar",
year = "2022",
abstract = "The objective of this research is to report results from a new ensemble method for
vegetation classification that uses deep learning (DL) and machine learning (ML)
techniques. Deep learning and machine learning architectures have recently been used
in methods for vegetation classification, proving their efficacy in several scientific
investigations. However, some limitations have been highlighted in the literature, such
as insufficient model variance and restricted generalization capabilities. Ensemble DL and
ML models has often been recommended as a feasible method to overcome these
constraints. A considerable increase in classification accuracy for vegetation classification
was achieved by growing an ensemble of decision trees and allowing them to vote for the
most popular class. An ensemble DL and ML architecture is presented in this study to
increase the prediction capability of individual DL and ML models. Three DL and ML
models, namely Convolutional Neural Network (CNN), Random Forest (RF), and biased
Support vector machine (B-SVM), are used to classify vegetation in the Eastern part of
Serbia, together with their ensemble form (CNN-RF-BSVM). The suggested DL and ML
ensemble architecture achieved the best modeling results with overall accuracy values
(0.93), followed by CNN (0.90), RF (0.91), and B-SVM (0.88). The results showed that the
suggested ensemble model outperformed the DL and ML models in terms of overall
accuracy by up to 5%, which was validated by the Wilcoxon signed-rank test. According to
this research, RF classifiers require fewer and easier-to-define user-defined parameters
than B-SVMs and CNN methods. According to overall accuracy analysis, the proposed
ensemble technique CNN-RF-BSVM also significantly improved classification
accuracy (by 4%).",
publisher = "Frontiers Media SA",
journal = "Frontiers in Environmental Science",
title = "Testing a New Ensemble Vegetation Classification Method Based on Deep Learning and Machine Learning Methods Using Aerial Photogrammetric Images",
volume = "10",
pages = "896158",
doi = "10.3389/fenvs.2022.896158"
}

Drobnjak, S., Stojanović, M., Djordjević, D., Bakrač, S., Jovanović, J.,& Djordjević, A.. (2022). Testing a New Ensemble Vegetation Classification Method Based on Deep Learning and Machine Learning Methods Using Aerial Photogrammetric Images. in Frontiers in Environmental Science
Frontiers Media SA., 10, 896158.
https://doi.org/10.3389/fenvs.2022.896158

Drobnjak S, Stojanović M, Djordjević D, Bakrač S, Jovanović J, Djordjević A. Testing a New Ensemble Vegetation Classification Method Based on Deep Learning and Machine Learning Methods Using Aerial Photogrammetric Images. in Frontiers in Environmental Science. 2022;10:896158.
doi:10.3389/fenvs.2022.896158 .

Drobnjak, Siniša, Stojanović, Marko, Djordjević, Dejan, Bakrač, Saša, Jovanović, Jasmina, Djordjević, Aleksandar, "Testing a New Ensemble Vegetation Classification Method Based on Deep Learning and Machine Learning Methods Using Aerial Photogrammetric Images" in Frontiers in Environmental Science, 10 (2022):896158,
https://doi.org/10.3389/fenvs.2022.896158 . .

TY  - CONF
AU  - Stojanović, Marko
AU  - Drobnjak, Siniša
AU  - Jovanović, Jasmina M.
AU  - Đorđević, Dejan
AU  - Banković, Radoje
PY  - 2021
UR  - http://gery.gef.bg.ac.rs/handle/123456789/1427
AB  - Kartografska generalizacija je kreativni proces apstrakcije, koji se koristi prilikom izrade karata.
Uključuje proučavanje geografskog okruženja, procenu i obradu geo-podataka u odnosu na vrstu, namenu i
razmeru karte. U eri digitalne kartografije veća pažnja se posvećuje automatizaciji procesa generalizacije i
razvoju alata za automatsko uopštavanje sadržaja. U ovom radu su objašnjeni modeli i procedure za izradu
zidne karte rezmere 1:250.000 (ZK250) na osnovu digitalne topografske karte razmere 1:250.000 (DTK250).
AB  - Cartographic generalization is a creative process of abstraction, which is used in the maps
production. It includes the study of the geographic environment, evaluation and processing of geodata, with
regard to type, purpose, and scale of the map. In the era of digital cartography, more attention is being paid
to the automatization of the genralization process and developing the tools for the automatic generalization
of content. This paper explains the models and procedures for the production of wall map at scale 1:250.000
(WM250) based on digital topographic map at scale 1:250.000 (DTM250).
PB  - Beograd : Univerzitet u Beogradu - Matematički fakultet
C3  - Zbornik radova XLVIII Simpozijum o operacionalnim istraživanjima "SYM-OP-IS 2021", Banja Koviljača
T1  - Kartografska generalizacija za potrebe izrade zidne karte razmere 1:250.000
T1  - Cartographic generalization for wall map production at scale 1:250.000
SP  - 171
EP  - 176
UR  - https://hdl.handle.net/21.15107/rcub_gery_1427
ER  - 

@conference{
author = "Stojanović, Marko and Drobnjak, Siniša and Jovanović, Jasmina M. and Đorđević, Dejan and Banković, Radoje",
year = "2021",
abstract = "Kartografska generalizacija je kreativni proces apstrakcije, koji se koristi prilikom izrade karata.
Uključuje proučavanje geografskog okruženja, procenu i obradu geo-podataka u odnosu na vrstu, namenu i
razmeru karte. U eri digitalne kartografije veća pažnja se posvećuje automatizaciji procesa generalizacije i
razvoju alata za automatsko uopštavanje sadržaja. U ovom radu su objašnjeni modeli i procedure za izradu
zidne karte rezmere 1:250.000 (ZK250) na osnovu digitalne topografske karte razmere 1:250.000 (DTK250)., Cartographic generalization is a creative process of abstraction, which is used in the maps
production. It includes the study of the geographic environment, evaluation and processing of geodata, with
regard to type, purpose, and scale of the map. In the era of digital cartography, more attention is being paid
to the automatization of the genralization process and developing the tools for the automatic generalization
of content. This paper explains the models and procedures for the production of wall map at scale 1:250.000
(WM250) based on digital topographic map at scale 1:250.000 (DTM250).",
publisher = "Beograd : Univerzitet u Beogradu - Matematički fakultet",
journal = "Zbornik radova XLVIII Simpozijum o operacionalnim istraživanjima "SYM-OP-IS 2021", Banja Koviljača",
title = "Kartografska generalizacija za potrebe izrade zidne karte razmere 1:250.000, Cartographic generalization for wall map production at scale 1:250.000",
pages = "171-176",
url = "https://hdl.handle.net/21.15107/rcub_gery_1427"
}

Stojanović, M., Drobnjak, S., Jovanović, J. M., Đorđević, D.,& Banković, R.. (2021). Kartografska generalizacija za potrebe izrade zidne karte razmere 1:250.000. in Zbornik radova XLVIII Simpozijum o operacionalnim istraživanjima "SYM-OP-IS 2021", Banja Koviljača
Beograd : Univerzitet u Beogradu - Matematički fakultet., 171-176.
https://hdl.handle.net/21.15107/rcub_gery_1427

Stojanović M, Drobnjak S, Jovanović JM, Đorđević D, Banković R. Kartografska generalizacija za potrebe izrade zidne karte razmere 1:250.000. in Zbornik radova XLVIII Simpozijum o operacionalnim istraživanjima "SYM-OP-IS 2021", Banja Koviljača. 2021;:171-176.
https://hdl.handle.net/21.15107/rcub_gery_1427 .

Stojanović, Marko, Drobnjak, Siniša, Jovanović, Jasmina M., Đorđević, Dejan, Banković, Radoje, "Kartografska generalizacija za potrebe izrade zidne karte razmere 1:250.000" in Zbornik radova XLVIII Simpozijum o operacionalnim istraživanjima "SYM-OP-IS 2021", Banja Koviljača (2021):171-176,
https://hdl.handle.net/21.15107/rcub_gery_1427 .