Spatio-Temporal Classification Framework for Mapping Woody Vegetation from Multi-Temporal Sentinel-2 Imagery
Само за регистроване кориснике
2020
Аутори
Kovačević, JovanCvijetinović, Željko
Lakušić, Dmitar
Kuzmanović, Nevena
Šinžar-Sekulić, Jasmina
Mitrović, Momir
Stančić, Nikola
Brodić, Nenad
Mihajlović, Dragan
Чланак у часопису (Објављена верзија)
Метаподаци
Приказ свих података о документуАпстракт
The inventory of woody vegetation is of great importance for good forest management. Advancements of remote sensing techniques have provided excellent tools for such purposes, reducing the required amount of time and labor, yet with high accuracy and the information richness. Sentinel-2 is one of the relatively new satellite missions, whose 13 spectral bands and short revisit time proved to be very useful when it comes to forest monitoring. In this study, the novel spatio-temporal classification framework for mapping woody vegetation from Sentinel-2 multitemporal data has been proposed. The used framework is based on the probability random forest classification, where temporal information is explicitly defined in the model. Because of this, several predictions are made for each pixel of the study area, which allow for specific spatio-temporal aggregation to be performed. The proposed methodology has been successfully applied for mapping eight potential forest and shrubby vegetation typ...es over the study area of Serbia. Several spatio-temporal aggregation approaches have been tested, divided into two main groups: pixel-based and neighborhood-based. The validation metrics show that determining the most common vegetation type classes in the neighborhood of 5 × 5 pixels provides the best results. The overall accuracy and kappa coefficient obtained from five-fold cross validation of the results are 82.97% and 0.75, respectively. The corresponding producer’s accuracies range from 36.74% to 97.99% and user’s accuracies range from 46.31% to 98.43%. The proposed methodology proved to be applicable for mapping woody vegetation in Serbia and shows a potential to be implemented in other areas as well. Further testing is necessary to confirm such assumptions.
Кључне речи:
classification / Sentinel-2 / woody vegetation / probability random forest / forest inventory / SerbiaИзвор:
Remote Sensing, 2020, 12, 17, 2845-Издавач:
- MDPI
Финансирање / пројекти:
- Унапређење геодетске инфраструктуре Србије за потребе савременог Државног премера (RS-MESTD-Technological Development (TD or TR)-36020)
- Биодиверзитет биљног света Србије и Балканског полуострва - процена, одрживо коришћење и заштита (RS-MESTD-Basic Research (BR or ON)-173030)
Колекције
Институција/група
GraFarTY - JOUR AU - Kovačević, Jovan AU - Cvijetinović, Željko AU - Lakušić, Dmitar AU - Kuzmanović, Nevena AU - Šinžar-Sekulić, Jasmina AU - Mitrović, Momir AU - Stančić, Nikola AU - Brodić, Nenad AU - Mihajlović, Dragan PY - 2020 UR - https://grafar.grf.bg.ac.rs/handle/123456789/2057 AB - The inventory of woody vegetation is of great importance for good forest management. Advancements of remote sensing techniques have provided excellent tools for such purposes, reducing the required amount of time and labor, yet with high accuracy and the information richness. Sentinel-2 is one of the relatively new satellite missions, whose 13 spectral bands and short revisit time proved to be very useful when it comes to forest monitoring. In this study, the novel spatio-temporal classification framework for mapping woody vegetation from Sentinel-2 multitemporal data has been proposed. The used framework is based on the probability random forest classification, where temporal information is explicitly defined in the model. Because of this, several predictions are made for each pixel of the study area, which allow for specific spatio-temporal aggregation to be performed. The proposed methodology has been successfully applied for mapping eight potential forest and shrubby vegetation types over the study area of Serbia. Several spatio-temporal aggregation approaches have been tested, divided into two main groups: pixel-based and neighborhood-based. The validation metrics show that determining the most common vegetation type classes in the neighborhood of 5 × 5 pixels provides the best results. The overall accuracy and kappa coefficient obtained from five-fold cross validation of the results are 82.97% and 0.75, respectively. The corresponding producer’s accuracies range from 36.74% to 97.99% and user’s accuracies range from 46.31% to 98.43%. The proposed methodology proved to be applicable for mapping woody vegetation in Serbia and shows a potential to be implemented in other areas as well. Further testing is necessary to confirm such assumptions. PB - MDPI T2 - Remote Sensing T1 - Spatio-Temporal Classification Framework for Mapping Woody Vegetation from Multi-Temporal Sentinel-2 Imagery IS - 17 SP - 2845 VL - 12 DO - https://doi.org/10.3390/rs12172845 ER -
@article{ author = "Kovačević, Jovan and Cvijetinović, Željko and Lakušić, Dmitar and Kuzmanović, Nevena and Šinžar-Sekulić, Jasmina and Mitrović, Momir and Stančić, Nikola and Brodić, Nenad and Mihajlović, Dragan", year = "2020", abstract = "The inventory of woody vegetation is of great importance for good forest management. Advancements of remote sensing techniques have provided excellent tools for such purposes, reducing the required amount of time and labor, yet with high accuracy and the information richness. Sentinel-2 is one of the relatively new satellite missions, whose 13 spectral bands and short revisit time proved to be very useful when it comes to forest monitoring. In this study, the novel spatio-temporal classification framework for mapping woody vegetation from Sentinel-2 multitemporal data has been proposed. The used framework is based on the probability random forest classification, where temporal information is explicitly defined in the model. Because of this, several predictions are made for each pixel of the study area, which allow for specific spatio-temporal aggregation to be performed. The proposed methodology has been successfully applied for mapping eight potential forest and shrubby vegetation types over the study area of Serbia. Several spatio-temporal aggregation approaches have been tested, divided into two main groups: pixel-based and neighborhood-based. The validation metrics show that determining the most common vegetation type classes in the neighborhood of 5 × 5 pixels provides the best results. The overall accuracy and kappa coefficient obtained from five-fold cross validation of the results are 82.97% and 0.75, respectively. The corresponding producer’s accuracies range from 36.74% to 97.99% and user’s accuracies range from 46.31% to 98.43%. The proposed methodology proved to be applicable for mapping woody vegetation in Serbia and shows a potential to be implemented in other areas as well. Further testing is necessary to confirm such assumptions.", publisher = "MDPI", journal = "Remote Sensing", title = "Spatio-Temporal Classification Framework for Mapping Woody Vegetation from Multi-Temporal Sentinel-2 Imagery", number = "17", pages = "2845", volume = "12", doi = "https://doi.org/10.3390/rs12172845" }
Kovačević, J., Cvijetinović, Ž., Lakušić, D., Kuzmanović, N., Šinžar-Sekulić, J., Mitrović, M., Stančić, N., Brodić, N.,& Mihajlović, D.. (2020). Spatio-Temporal Classification Framework for Mapping Woody Vegetation from Multi-Temporal Sentinel-2 Imagery. in Remote Sensing MDPI., 12(17), 2845. https://doi.org/https://doi.org/10.3390/rs12172845
Kovačević J, Cvijetinović Ž, Lakušić D, Kuzmanović N, Šinžar-Sekulić J, Mitrović M, Stančić N, Brodić N, Mihajlović D. Spatio-Temporal Classification Framework for Mapping Woody Vegetation from Multi-Temporal Sentinel-2 Imagery. in Remote Sensing. 2020;12(17):2845. doi:https://doi.org/10.3390/rs12172845 .
Kovačević, Jovan, Cvijetinović, Željko, Lakušić, Dmitar, Kuzmanović, Nevena, Šinžar-Sekulić, Jasmina, Mitrović, Momir, Stančić, Nikola, Brodić, Nenad, Mihajlović, Dragan, "Spatio-Temporal Classification Framework for Mapping Woody Vegetation from Multi-Temporal Sentinel-2 Imagery" in Remote Sensing, 12, no. 17 (2020):2845, https://doi.org/https://doi.org/10.3390/rs12172845 . .