跳到主要內容

簡易檢索 / 詳目顯示

回結果列表
研究生： 潘易翰
Ilham Adi Panuntun
論文名稱： 建議的 LSST-Former 深度學習架構基於少樣本學習，用於小資料集的紅樹林損耗檢測
Proposed LSST-Former Deep Learning Architecture based on Few-Shot Learning for Mangrove Loss Detection with a Small Dataset
指導教授： 陳映濃
Chen, Ying-Nong
口試委員:
學位類別： 碩士
Master
系所名稱： 太空及遙測研究中心 - 遙測科技碩士學位學程
Master of Science Program in Remote Sensing Science and Technology
論文出版年： 2024
畢業學年度： 112
語文別： 英文
論文頁數： 59
中文關鍵詞： 紅樹林喪失檢測 、少樣本學習 、Transformer 、全卷積網絡 (FCN)
外文關鍵詞： mangrove loss detection, few-shot learning, Transformer, Fully Convolutional Network (FCN)
相關次數： 點閱：18  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 紅樹林是提供各種生態和社會經濟效益的關鍵生態系統，但它們受到森林砍伐和城市化等人類活動的威脅。傳統的紅樹林損失監測方法依賴於勞動密集和耗時的現場調查或高解析度衛星圖像分析，通常在空間覆蓋和時間分辨率上存在限制。 LSST-Former架構整合了FCN、基於Transformer的模型和少樣本學習技術的優勢，以應對使用小數據集進行紅樹林損失檢測的挑戰。 Transformer已經在捕捉長程依賴性和從序列數據中學習方面取得了顯著成功，而少樣本學習使模型能夠很好地對未見過的類別或任務進行泛化，並且具有有限的訓練示例。通過結合這些方法，LSST-Former旨在有效地從異構類別中學習。我們的實驗結果展示了LSST-Former相對於現有的深度學習架構（如隨機森林、支援向量機、U-Net、LinkNet、Vision Transformer、SpectralFormer、MDPrePost-Net 和SST-Former）的優越性能，凸顯了其在紅樹林保護和管理工作中實際應用的潛力。

    Mangroves are crucial ecosystems that provide various ecological and socio-economic benefits, but they are under threat from anthropogenic activities such as deforestation and urbanization. Traditional methods for monitoring mangrove loss rely on labor-intensive and time-consuming field surveys or high-resolution satellite imagery analysis, which are often limited in spatial coverage and temporal resolution. The LSST-Former architecture integrates the strengths of both FCN, Transformer-based models, and few-shot learning techniques to address the challenges of mangrove loss detection with small datasets. Transformers have demonstrated remarkable success in capturing long-range dependencies and learning from sequential data, while few-shot learning enables models to generalize well to unseen classes or tasks with limited training examples. By combining these approaches, LSST-Former aims to learn from heterogeneous classes effectively. Our experimental results showcase the superior performance of LSST-Former compared to existing deep learning architectures such as random forest, Support Vector Machine, U-Net, LinkNet, Vision Transformer, SpectralFormer, MDPrePost-Net, and SST-Former, highlighting its potential for practical applications in mangrove conservation and management efforts.

    摘要 i Abstract ii Acknowledgement iii Table of Contents iv List of Figures vi List of Tables vii CHAPTER I INTRODUCTION 1 1.1. Background 1 1.2. Challenge and Objectives 3 CHAPTER II LITERATURE REVIEW 4 2.1. Mangroves 4 2.2. Mangrove Loss 5 2.3. Remote Sensing for Mangrove 5 2.4. Deep Learning for Remote Sensing 6 2.5. Related Works 7 CHAPTER III STUDY AREA AND METHODS 8 3.1. Study Area 8 3.2. Datasets 9 3.2.1. Sentinel-2 Pre-Processing 9 3.2.2. Input Data for Model 12 3.3. Methods 14 3.4. Proposed Deep Learning Architecture 15 3.4.1. FCN Architecture 15 3.4.2. Transformer Architecture 17 3.5. Evaluation Assessment 21 3.6. Validation of Universal Applicability Model 22 3.7. Implementation Detail 23 CHAPTER IV RESULTS AND DISCUSSION 24 4.1. Proposed Deep Learning Architecture Result 24 4.2. Ablation Experiments 25 4.2.1. The influence of different training sizes 25 4.2.2. Effects of Each Extractor Part 26 4.2.3. The Impact of Mangrove and Vegetation Indices 26 4.2.4. Effects of Parameters 27 4.3. Examine the Contrast with Other Well-Established Architectures 28 4.4. Universal Applicability of the Model 30 4.5. Discussion 33 CHAPTER V CONCLUSIONS AND FUTURE WORK 40 5.1. Conclusions 40 5.2. Future Work 40 References 41

    [1] FAO, U. The world’s mangroves 1980–2005. FAO forestry paper 2007, 153, 77.
    [2] Ahmed, S.; Kamruzzaman, M.; Rahman, M.S.; Sakib, N.; Azad, M.S.; Dey, T. Stand structure and carbon storage of a young mangrove plantation forest in coastal area of Bangladesh: the promise of a natural solution. Nature-Based Solutions 2022, 2, 100025.
    [3] Sunkur, R.; Kantamaneni, K.; Bokhoree, C.; Ravan, S. Mangroves’ Role in Supporting Ecosystem-Based Techniques to Reduce Disaster Risk and Adapt to Climate Change: A Review. Journal of Sea Research 2023, 196, 102449.
    [4] Yudha, R.P.; Sugito, Y.S.; Sillanpää, M.; Nurvianto, S. Impact of logging on the biodiversity and composition of flora and fauna in the mangrove forests of Bintuni Bay, West Papua, Indonesia. Forest Ecology and Management 2021, 488, 119038.
    [5] Lopes, R.G.P.S.; Rego, A.P.; de Jesus Gomes, S.M.; Antonio, Í.G.; Freire, T.B.; Coimbra, M.R.M. Effects of salinity on pre-and post-fertilization developmental events in the mangrove oyster Crassostrea rhizophorae (GUILDING, 1828). Theriogenology 2024, 218, 62-68.
    [6] Richards, D.R.; Friess, D.A. Rates and drivers of mangrove deforestation in Southeast Asia, 2000–2012. Proceedings of the National Academy of Sciences 2016, 113, 344-349.
    [7] Gitau, P.N.; Duvail, S.; Verschuren, D. Evaluating the combined impacts of hydrological change, coastal dynamics and human activity on mangrove cover and health in the Tana River delta, Kenya. Regional Studies in Marine Science 2023, 61, 102898.
    [8] Quevedo, J.M.D.; Lukman, K.M.; Ulumuddin, Y.I.; Uchiyama, Y.; Kohsaka, R. Applying the DPSIR Framework to Qualitatively Assess the Globally Important Mangrove Ecosystems of Indonesia: A Review towards Evidence-Based Policymaking Approaches. Marine Policy 2023, 147, 105354.
    [9] Chopade, M.R.; Mahajan, S.; Chaube, N. Assessment of land use, land cover change in the mangrove forest of Ghogha area, Gulf of Khambhat, Gujarat. Expert Systems with Applications 2023, 212, 118839.
    [10] Kudrass, H.R.; Hanebuth, T.J.J.; Zander, A.M.; Linstädter, J.; Akther, S.H.; Shohrab, U.M. Architecture and Function of Salt-Producing Kilns from the 8th to 18th Century in the Coastal Sundarbans Mangrove Forest, Central Ganges-Brahmaputra Delta, Bangladesh. Archaeological Research in Asia 2022, 32, 100412.
    [11] Vizcaya-Martínez, D.A.; Flores-de-Santiago, F.; Valderrama-Landeros, L.; Serrano, D.; Rodríguez-Sobreyra, R.; Álvarez-Sánchez, L.F.; Flores-Verdugo, F. Monitoring Detailed Mangrove Hurricane Damage and Early Recovery Using Multisource Remote Sensing Data. Journal of Environmental Management 2022, 320, 115830.
    [12] Ward, R.D.; Drude de Lacerda, L. Responses of mangrove ecosystems to sea level change. In Dynamic Sedimentary Environments of Mangrove Coasts; Elsevier: 2021; pp. 235-253.
    [13] Gomes, L.E.d.O.; Sanders, C.J.; Nobrega, G.N.; Vescovi, L.C.; Queiroz, H.M.; Kauffman, J.B.; Ferreira, T.O.; Bernardino, A.F. Ecosystem Carbon Losses Following a Climate-Induced Mangrove Mortality in Brazil. Journal of Environmental Management 2021, 297, 113381.
    [14] Padonou, E.A.; Gbaï, N.I.; Kolawolé, M.A.; Idohou, R.; Toyi, M. How far are mangrove ecosystems in Benin (West Africa) conserved by the Ramsar Convention? Land Use Policy 2021, 108, 105583.
    [15] Tengku Hashim, T.M.Z.; Engku Ariff, E.A.R.; Suratman, M.N. Aquaculture in mangroves. Mangroves: Ecology, Biodiversity and Management 2021, 419-438.
    [16] Sahriman, N.; Zainal, M.Z.; Ghazali, N.; Abbas, M.A. Appraising effect of environmental parameter toward mangrove area: A review. In Proceedings of the 2017 IEEE 13th International Colloquium on Signal Processing & its Applications (CSPA), 2017; pp. 287-292.
    [17] Phan, M.H.; Stive, M.J. Managing mangroves and coastal land cover in the Mekong Delta. Ocean & Coastal Management 2022, 219, 106013.
    [18] Valiela, I.; Bowen, J.L.; York, J.K. Mangrove Forests: One of the World’s Threatened Major Tropical Environments. Bioscience 2001, 51, 807–815.
    [19] Goldberg, L.; Lagomasino, D.; Thomas, N.; Fatoyinbo, T. Global Declines in Human-driven Mangrove Loss. Glob. Chang. Biol. 2020, 26, 5844–5855.
    [20] Hamilton, S.E.; Casey, D. Creation of a High Spatio-Temporal Resolution Global Database of Continuous Mangrove Forest Cover for the 21st Century (CGMFC-21). Glob. Ecol. Biogeogr. 2016, 25, 729–738.
    [21] Lin, C.H.; Chu, M.C.; Tang, P.W. CODE-MM: Convex Deep Mangrove Mapping Algorithm Based On Optical Satellite Images. IEEE Transactions on Geoscience and Remote Sensing 2023.
    [22] Jamaluddin, I.; Thaipisutikul, T.; Chen, Y.N.; Chuang, C.H.; Hu, C.L. MDPrePost-Net: A Spatial-Spectral-Temporal Fully Convolutional Network for Mapping of Mangrove Degradation Affected by Hurricane Irma 2017 Using Sentinel-2 Data. Remote Sensing 2021, 13, 5042.
    [23] Quang, N.H.; Quinn, C.H.; Carrie, R.; Stringer, L.C.; Hackney, C.R.; Van Tan, D. Comparisons of regression and machine learning methods for estimating mangrove above-ground biomass using multiple remote sensing data in the red River Estuaries of Vietnam. Remote Sensing Applications: Society and Environment 2022, 26, 100725.
    [24] de Souza Moreno, G.M.; de Carvalho Júnior, O.A.; de Carvalho, O.L.F.; Andrade, T.C. Deep semantic segmentation of mangroves in Brazil combining spatial, temporal, and polarization data from Sentinel-1 time series. Ocean & Coastal Management 2023, 231, 106381.
    [25] Li, L.; Zhang, W.; Zhang, X.; Emam, M.; Jing, W. Semi-Supervised Remote Sensing Image Semantic Segmentation Method Based on Deep Learning. Electronics 2023, 12, 348.
    [26] Wang, Y.; Hong, D.; Sha, J.; Gao, L.; Liu, L.; Zhang, Y.; Rong, X. Spectral–Spatial–Temporal Transformers for Hyperspectral Image Change Detection. IEEE Transactions on Geoscience and Remote Sensing 2022, 60, 1–14.
    [27] Lugo, A.E.; Snedaker, S.C. The ecology of mangroves. Annual review of ecology and systematics 1974, 5, 39-64.
    [28] Lugo, A.E.; Sell, M.; Snedaker, S.C. Mangrove ecosystem analysis. Systems analysis and simulation in ecology 1976, 4, 113-145.
    [29] Biswas, S.R.; Mallik, A.U.; Choudhury, J.K.; Nishat, A. A unified framework for the restoration of Southeast Asian mangroves—bridging ecology, society and economics. Wetlands Ecology and Management 2009, 17, 365-383.
    [30] Kmarius. Mangroves Trees Rocks. Available online: https://pixabay.com/photos/mangroves-trees-rocks-river-nature-5205415/ (accessed on 13 April 2024).
    [31] Lillesand, T.; Kiefer, R.W.; Chipman, J. Remote sensing and image interpretation; John Wiley & Sons: 2015.
    [32] Huete, A.R.; Glenn, E.P. Remote sensing of ecosystem structure and function. Advances in Environmental Remote Sensing. Sensors, Algorithms, and Applications. CRC Press, Boca Raton, Florida, USA 2011, 291-320.
    [33] Swain, P.H.; Davis, S.M. Remote sensing: the quantitative approach. IEEE Transactions on Pattern Analysis & Machine Intelligence 1981, 3, 713-714.
    [34] Roman, A.; Ursu, T. Multispectral satellite imagery and airborne laser scanning techniques for the detection of archaeological vegetation marks. In Landscape archaeology on the northern frontier of the roman empire at porolissum: an interdisciplinary research project. Cluj-Napoca: Mega Publishing House; 2016; pp. 141-152.
    [35] Liu, Y.; Hu, J.; Kang, X.; Luo, J.; Fan, S. Interactformer: Interactive Transformer and CNN for Hyperspectral Image Super-Resolution. IEEE Transactions on Geoscience and Remote Sensing 2022, 60, 1–15.
    [36] Sun, L.; Zhao, G.; Zheng, Y.; Wu, Z. Spectral–Spatial Feature Tokenization Transformer for Hyperspectral Image Classification. IEEE Transactions on Geoscience and Remote Sensing 2022, 60, 1-14.
    [37] Hong, D.; Han, Z.; Yao, J.; Gao, L.; Zhang, B.; Plaza, A.; Chanussot, J. SpectralFormer: Rethinking Hyperspectral Image Classification With Transformers. IEEE Transactions on Geoscience and Remote Sensing 2022, 60, 1–15.
    [38] Shelhamer, E.; Long, J.; Darrell, T. Fully Convolutional Networks for Semantic Segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence 2017, 39, 640-651.
    [39] Ronneberger, O.; Fischer, P.; Brox, T. U-Net: Convolutional Networks for Biomedical Image Segmentation. Proceedings of the Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015 2015, 234–241.
    [40] Chaurasia, A.; Culurciello, E. LinkNet: Exploiting Encoder Representations for Efficient Semantic Segmentation. In Proceedings of the Proceedings of the 2017 IEEE Visual Communications and Image Processing (VCIP), St. Petersburg, FL, USA, 2017, 2017/12/10-2017/12/13.
    [41] Saferbekov, S.; Iglovikov, V.; Buslaev, A.; Shvets, A. Feature Pyramid Network for Multi-Class Land Segmentation. Comput. Vis. Pattern Recognit. 2018, 2, 272-2723.
    [42] Zhao, H.; Shi, J.; Qi, X.; Wang, X.; Jia, J. Pyramid Scene Parsing Network. In Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017.
    [43] Panuntun, I.A.; Jamaluddin, I.; Chen, Y.-N.; Lai, S.-N.; Fan, K.-C. LinkNet-Spectral-Spatial-Temporal Transformer Based on Few-Shot Learning for Mangrove Loss Detection with Small Dataset. Remote Sensing 2024, 16, 1078.
    [44] Hamilton, S.E.; Friess, D.A. Global Carbon Stocks and Potential Emissions Due to Mangrove Deforestation from 2000 to 2012. Nature Climate Change 2018, 8, 240-244.
    [45] Masiyah, S.; Aryani, W. The evaluate soil quality of mangrove forest in Merauke, Papua. In Proceedings of the IOP Conference Series: Earth and Environmental Science, 2021; p. 012019.
    [46] Rumondang, A.L.; Kusmana, C.; Budi, S.W. Species Composition and Structure of Angke Kapuk Mangrove Protected Forest, Jakarta, Indonesia. Biodiversitas Journal of Biological Diversity 2021, 22, 9.
    [47] Liu, C.C.; Hsu, T.W.; Wen, H.L.; Wang, K.H. Mapping Pure Mangrove Patches in Small Corridors and Sandbanks Using Airborne Hyperspectral Imagery. Remote Sens. 2019, 11, 592.
    [48] ESRI | World Imagery Wayback. Available online: https://livingatlas.arcgis.com/wayback/ (accessed on 08 December 2023).
    [49] Google Earth Engine. Available online: https://developers.google.com/earth-engine/datasets/catalog/sentinel-2 (accessed on 04 November 2023).
    [50] Yin, F.; Lewis, P.E.; Gómez-Dans, J.L. Bayesian Atmospheric Correction over Land: Sentinel-2/MSI and Landsat 8/OLI. 2022.
    [51] Diniz, C.; Cortinhas, L.; Nerino, G.; Rodrigues, J.; Sadeck, L.; Adami, M.; Souza-Filho, P. Brazilian Mangrove Status: Three Decades of Satellite Data Analysis. Remote Sensing 2019, 11, 808.
    [52] Tucker, C.J. Red and photographic infrared linear combinations for monitoring vegetation. Remote Sens. Environ. 1979, 8, 127-150.
    [53] Gupta, K.; Mukhopadhyay, A.; Giri, S.; Chanda, A.; Datta Majumdar, S.; Samanta, S.; Mitra, D.; Samal, R.N.; Pattnaik, A.K.; Hazra, S. An index for discrimination of mangroves from non-mangroves using LANDSAT 8 OLI imagery. MethodsX 2018, 5, 1129-1139.
    [54] McFeeters, S.K. The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. Int. J. Remote Sens. 1996, 17, 1425-1432.
    [55] Shi, T.; Liu, J.; Hu, Z.; Liu, H.; Wang, J.; Wu, G. New spectral metrics for mangrove forest identification. Remote Sens. Lett. 2016, 7, 885-894.
    [56] Xu, H. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. Int. J. Remote Sens. 2006, 27, 3025-3033.
    [57] Szegedy, C.; Vanhoucke, V.; Ioffe, S.; Shlens, J.; Wojna, Z. Rethinking the Inception Architecture for Computer Vision. In Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.
    [58] Vaswani, A.; Shazeer, N.; Parmar, N.; Uszkoreit, J.; Jones, L.; Gomez, A.N.; Kaiser, Ł.; Polosukhin, I. Attention is all you need. In Proceedings of the Advances in neural information processing systems, 2017.
    [59] Gehring, J.; Auli, M.; Grangier, D.; Yarats, D.; Dauphin, Y.N. Convolutional sequence to sequence learning. In Proceedings of the International conference on machine learning, 2017; pp. 1243-1252.
    [60] Ren, M.; Triantafillou, E.; Ravi, S.; Snell, J.; Swersky, K.; Tenenbaum, J.; Larochelle, H.; Zemel, R. Meta-Learning for Semi-Supervised Few-Shot Classification. arXiv 2018, arXiv:1803.00676.
    [61] Cheng, G., Yan, B., Shi, P., Li, K., Yao, X., Guo, L., & Han, J. (2021). Prototype-CNN for few-shot object detection in remote sensing images. IEEE Transactions on Geoscience and Remote Sensing, 60, 1-10.
    [62] Rezatofighi, S.H.; Tsoi, N.; Gwak, J.; Sadeghian, A.; Reid, I.D.; Savarese, S. Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression. In Proceedings of the Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA, 2019/06/15-2019/06/20, 2019.
    [63] Sokolova, M.; Japkowicz, N.; Szpakowicz, S. Beyond Accuracy, F-Score and ROC: A Family of Discriminant Measures for Performance Evaluation. AI 2006: Advances in Artificial Intelligence 2006, 4304, 1015-1021.
    [64] Dosovitskiy, A.; Beyer, L.; Kolesnikov, A.; Weissenborn, D.; Zhai, X.; Unterthiner, T.; Dehghani, M.; Minderer, M.; Heigold, G.; Gelly, S.; Uszkoreit, J. An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929 2020, 22 October 2020.
    [65] Reflected Near-Infrared Waves. Available online: https://science.nasa.gov/ems/08_nearinfraredwaves/ (accessed on 9 February 2024).

    QR CODE
    :::