Ứng dụng eCognition phân loại thảm phủ rừng phòng hộ Dầu Tiếng, tỉnh Tây Ninh
Article Sidebar
Full Text: 
PDF
Abstract
This study presents the application of eCognition software to classify forest status in the Dau Tieng Protection Forest Management Board, Tay Ninh Province, using Sentinel-2A satellite images. The land cover classification method using tools such as “Assign Class” and “Classification” in eCognition and field survey data from 80 key image interpretation points, combined with standard plots, as well as a rule-based classification method using logical thresholds that does not require training data (Rule Set) were used in this study. The results show that the area was classified into eight land cover classes, including five forest types, with a total forested area of 25.156.53 hectares (accounting for 89.33%). These forest types include: medium evergreen forest, poor evergreen forest, very poor evergreen forest, recovery forest and planted forest, with the classification accuracy reaching 82.5% and the Kappa coefficient is 0.8. The study also shows that the use of the "Classification" tool in classifying forest objects has higher accuracy than classification based on the threshold value of spectrum indicators (Rule Set).
Tóm tắt
Nghiên cứu này được thực hiện nhằm trình bày ứng dụng phần mềm eCognition để phân loại thảm phủ rừng phòng hộ tại Dầu Tiếng, tỉnh Tây Ninh bằng ảnh vệ tinh Sentinel-2A. Phương pháp phân loại thảm phủ sử dụng các công cụ “Assign Class”, “Classification” trong phần mềm eCognition kết hợp dữ liệu điều kiện ngoại nghiệp từ 80 mẫu khóa ảnh, các ô tiêu chuẩn và phân loại theo phương pháp huấn luyện theo ngưỡng logic không cần huấn luyện (Rule Set) đã được sử dụng trong nghiên cứu này. Kết quả cho thấy tại khu vực được phân loại thành 8 lớp phủ, trong đó có 5 trạng thái rừng với tổng diện tích có rừng là 25.156,53 ha (chiếm 89,33%) bao gồm: rừng thường xanh trung bình, thường xanh nghèo, thường xanh kiệt, thường xanh phục hồi và rừng trồng với độ chính xác phân loại đạt 82,5% và hệ số Kappa là 0,8. Kết quả nghiên cứu cũng cho thấy việc sử dụng công cụ “Classification” trong phân loại đối tượng rừng có độ chính xác và hiệu quả cao hơn so với phân loại dựa trên ngưỡng giá trị của các chỉ số phổ (Rule Set).
Article Details
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Tài liệu tham khảo
Aksoy, S., & Akçay, H. G. (2005). Multi-resolution segmentation and shape analysis for remote sensing image classification.
Awate, P., & Nagne, A. (2023). Enhanced Vegetation Cover Assessment Using Sentinel-2A: A Unified Perspective on NDVI, SAVI, and GNDVI.
Blaschke, T. (2010). Object based image analysis for remote sensing. ISPRS Journal of Photogrammetry and Remote Sensing, 65(1), 2-16. doi:https://doi.org/10.1016/j.isprsjprs.2009.06.004
Bui, H. M., Nguyen, P. T. B., & Nguyen, T. T. (2021). Variation in vegetation cover and its relationship with natural and social factors in Yen Chau, Son La. Journal of Forestry Science and Technology, 3, 031-040 (in Vietnamese).
Duveiller, G., Defourny, P., Desclée, B., & Mayaux, P. (2008). Deforestation in Central Africa: Estimates at regional, national and landscape levels by advanced processing of systematically-distributed Landsat extracts. Remote sensing of environment, 112(5), 1969-1981.
Forkuor, G., Dimobe, K., Serme, I., & Tondoh, J. E. (2018). Landsat-8 vs. Sentinel-2: examining the added value of sentinel-2’s red-edge bands to land-use and land-cover mapping in Burkina Faso. GIScience & remote sensing, 55(3), 331-354. doi:https://doi.org/10.1080/15481603.2017.1370169
Guimarães, N., Pádua, L., Marques, P., Silva, N., Peres, E., & Sousa, J. J. (2020). Forestry remote sensing from unmanned aerial vehicles: A review focusing on the data, processing and potentialities. Remote Sensing, 12(6), 1046. doi:https://doi.org/10.3390/rs12061046
Kairu, E. (1982). An introduction to remote sensing. GeoJournal, 6, 251-260. doi:10.1007/BF00210657
Laliberte, A. S., Fredrickson, E. L., & Rango, A. (2007). Combining decision trees with hierarchical object-oriented image analysis for mapping arid rangelands. Photogrammetric engineering & Remote sensing, 73(2), 197-207. doi:https://doi.org/10.14358/PERS.73.2.197
Lambin, E. F. (1997). Modelling and monitoring land-cover change processes in tropical regions. Progress in physical geography, 21(3), 375-393. doi:https://doi.org/10.1177/030913339702100303
Landis, J., & Koch, G. (1977). The Measurement Of Observer Agreement For Categorical Data. Biometrics, 33, 159-174. doi:10.2307/2529310
Lapinskas, A. A. (2023). Influence of mining rent on the efficiency of using natural potential: the paradox of plenty and its Russian specifics. Записки Горного института(259 (eng)), 79-94. doi:10.31897/PMI.2023.13
Le, H. Q., & Bui, H. M (2018). Using Landsat remote sensing images and GIS to build a map of forest area changes in Doan Hung District, Phu Tho Province. Journal of Forestry Science, 4, 127-137 (in Vietnamese).
Lourenço, P., Teodoro, A. C., Gonçalves, J. A., Honrado, J. P., Cunha, M., & Sillero, N. (2021). Assessing the performance of different OBIA software approaches for mapping invasive alien plants along roads with remote sensing data. International Journal of Applied Earth Observation and Geoinformation, 95, 102263. doi:https://doi.org/10.1016/j.jag.2020.102263
Luu, C., Pham, B. T., Van Phong, T., Costache, R., Nguyen, H. D., Amiri, M., & Prakash, I. (2021). GIS-based ensemble computational models for flood susceptibility prediction in the Quang Binh Province, Vietnam. Journal of Hydrology, 599, 126500. doi:https://doi.org/10.1016/j.jhydrol.2021.12650
Maselli, F. (2004). Monitoring forest conditions in a protected Mediterranean coastal area by the analysis of multiyear NDVI data. Remote sensing of environment, 89(4), 423-433. doi:https://doi.org/10.1016/j.rse.2003.10.020
Nguyen, T. V., & T, B. Q. (2014). Application of Object-Oriented Image Classification Technique to Classify Forest Status According to Circular No. 34. Journal of Forestry Science, 2, 3343-3353 (in Vietnamese).
Nguyen, H. H., & Nguyen, Q. V. (2017). Using Landsat remote sensing images and GIS to build a map of forest area changes in the buffer zone of Xuan Son National Park. Journal of Forestry Science and Technology, 3, 046-056 (in Vietnamese).
Nguyen, C. T., Nguyen, H. H., & Tran, B. Q. (2019a). Establishing a mangrove forest map in 2018 of Thai Binh province from Sentinel 2A image data. Journal of Forestry Science and Technology, 6, 57 - 66 (in Vietnamese).
Nguyen, K, M., Nguyen, T. T. L., Bui, P. K., & Tran, D. L. H. (2019b). Application of remote sensing to assess changes in forest resources: A typical case in Chu Prong district, Gia Lai province. Journal of Science, An Giang University, 22(1), 67–80 (in Vietnamese).
Oreti, L., Giuliarelli, D., Tomao, A., & Barbati, A. (2021). Object oriented classification for mapping mixed and pure forest stands using very-high resolution imagery. Remote Sensing, 13(13), 2508. doi:https://doi.org/10.3390/rs13132508
Phiri, D., Simwanda, M., Nyirenda, V., Murayama, Y., & Ranagalage, M. (2020). Decision tree algorithms for developing rulesets for object-based land cover classification. ISPRS International Journal of Geo-Information, 9(5), 329. doi:https://doi.org/10.3390/ijgi9050329
Rogan, J., Franklin, J., & Roberts, D. A. (2002). A comparison of methods for monitoring multitemporal vegetation change using Thematic Mapper imagery. Remote sensing of environment, 80(1), 143-156. doi:https://doi.org/10.1016/S0034-4257(01)00296-6
Smirnyakov, V. V., Rodionov, V. A., Smirnyakova, V. V., & Orlov, F. A. (2022). The influence of the shape and size of dust fractions on their distribution and accumulation in mine workings when changing the structure of air flow. Записки Горного института, 253, 71-81. doi:10.31897/PMI.2022.12.
Tran, B, Q., Nguyen, L. D., & La, K. N. (2017). Application of GIS and remote sensing in analyzing the current status and assessing changes in forest resources in Vinh Cuu district, Dong Nai province. Journal of Forestry Science and Technology, 6, 092–100 (in Vietnamese).
Tran, B. Q., Cao, V. L. Q., Vo, H. M., & Nguyen, H. T. (2022). Building a forest status map of Dong Phu district, Binh Phuoc province using Sentinel 2a satellite images. Journal of Forestry Science and Technology, 3, 053-063 (in Vietnamese). doi:https://doi.org/10.55250/jo.vnuf.2022.3.053-063.