딥러닝과 Landsat 8 영상을 이용한 캘리포니아 산불 피해지 탐지 = Detection of Wildfire Burned Areas in California Using Deep Learning and Landsat 8 Images

다국어 초록 (Multilingual Abstract)

The increasing frequency of wildfires due to climate change is causing extreme loss of lifeand property. They cause loss of vegetation and affect ecosystem changes depending on their intensityand occurrence. Ecosystem changes, in turn, affect wildfire occurrence, causing secondary damage. Thus,accurate estimation of the areas affected by wildfires is fundamental. Satellite remote sensing is used forforest fire detection because it can rapidly acquire topographic and meteorological information aboutthe affected area after forest fires. In addition, deep learning algorithms such as convolutional neuralnetworks (CNN) and transformer models show high performance for more accurate monitoring of fireburntregions. To date, the application of deep learning models has been limited, and there is a scarcityof reports providing quantitative performance evaluations for practical field utilization. Hence, this studyemphasizes a comparative analysis, exploring performance enhancements achieved through both modelselection and data design. This study examined deep learning models for detecting wildfire-damagedareas using Landsat 8 satellite images in California. Also, we conducted a comprehensive comparisonand analysis of the detection performance of multiple models, such as U-Net and High-ResolutionNetwork-Object Contextual Representation (HRNet-OCR). Wildfire-related spectral indices such asnormalized difference vegetation index (NDVI) and normalized burn ratio (NBR) were used as inputchannels for the deep learning models to reflect the degree of vegetation cover and surface moisturecontent. As a result, the mean intersection over union (mIoU) was 0.831 for U-Net and 0.848 for HRNet-OCR, showing high segmentation performance. The inclusion of spectral indices alongside the base wavelength bands resulted in increased metric values for all combinations, affirming that the augmentationof input data with spectral indices contributes to the refinement of pixels. This study can be applied toother satellite images to build a recovery strategy for fire-burnt areas.

참고문헌 (Reference)

1 염종민 ; 한경수 ; 김인환, "장기간 SPOT/VEGETATION 정규화 식생지수를 이용한 지면 변화 탐지 개선에 관한 연구" 한국지리정보학회 13 (13): 111-124, 2010

2 김서연 ; 윤유정 ; 정예민 ; 권춘근 ; 서경원 ; 이양원, "위성기반 산불피해지수를 이용한 북한지역 산불피해지 분석" 대한원격탐사학회 38 (38): 1861-1869, 2022

3 최승필 ; 김동희 ; 建石陸太郞, "식생의 분광 반사특성을 이용한 산불 피해지 분석" 대한공간정보학회 14 (14): 89-94, 2006

4 차성은 ; 원명수 ; 장근창 ; 김경민 ; 김원국 ; 백승일 ; 임중빈, "농림위성 활용을 위한 산불 피해지 분류 딥러닝 알고리즘 평가" 대한원격탐사학회 38 (38): 1273-1283, 2022

5 윤형진 ; 정종철, "기계학습을 이용한 Sentinel-2 산불피해등급 분류" 국토연구원 106 : 107-117, 2020

6 박성욱 ; 김형우 ; 이수진 ; 윤예슬 ; 김은숙 ; 임종환 ; 이양원, "고해상도 위성영상과 Fully Convolutional Network를 활용한 산림재해 피해지 탐지" 한국사진지리학회 28 (28): 87-101, 2018

7 Luke Collins, "Training data requirements for fire severity mapping using Landsat imagery and random forest" Elsevier BV 245 : 111839-, 2020

8 Yushi Chen, "Spectral–Spatial Classification of Hyperspectral Data Based on Deep Belief Network" Institute of Electrical and Electronics Engineers (IEEE) 8 (8): 2381-2392, 2015

9 Magdalini Pleniou, "Sensitivity of spectral reflectance values to different burn and vegetation ratios: A multi-scale approach applied in a fire affected area" Elsevier BV 79 : 199-210, 2013

10 T. Loboda, "Regionally adaptable dNBR-based algorithm for burned area mapping from MODIS data" Elsevier BV 109 (109): 429-442, 2007

11 Ronneberger, O, "Medical image computing and computer-assisted intervention – MICCAI 2015" Springer 234-241, 2015

12 Zhang, Y., "Medical image computing and computer assisted intervention – MICCAI 2021" Springer 14-24, 2021

13 D.P. Roy, "Landsat-8: Science and product vision for terrestrial global change research" Elsevier BV 145 : 154-172, 2014

14 Riyanti Djalante, "Key assessments from the IPCC special report on global warming of 1.5 °C and the implications for the Sendai framework for disaster risk reduction" Elsevier BV 1 : 100001-, 2019

15 Emilio Chuvieco, "Historical background and current developments for mapping burned area from satellite Earth observation" Elsevier BV 225 : 45-64, 2019

16 Tsung-Yi Lin, "Focal Loss for Dense Object Detection" Institute of Electrical and Electronics Engineers (IEEE) 42 (42): 318-327, 2020

17 S. Escuin, "Fire severity assessment by using NBR (Normalized Burn Ratio) and NDVI (Normalized Difference Vegetation Index) derived from LANDSAT TM/ETM images" Informa UK Limited 29 (29): 1053-1073, 2007

18 Kendra K. McLauchlan, "Fire as a fundamental ecological process: Research advances and frontiers" Wiley 108 (108): 2047-2069, 2020

19 Fernando Carvajal-Ramírez, "Evaluation of Fire Severity Indices Based on Pre- and Post-Fire Multispectral Imagery Sensed from UAV" MDPI AG 11 (11): 993-, 2019

20 Lei Ma, "Deep learning in remote sensing applications: A meta-analysis and review" Elsevier BV 152 : 166-177, 2019

21 Xiao Xiang Zhu, "Deep Learning in Remote Sensing: A Comprehensive Review and List of Resources" Institute of Electrical and Electronics Engineers (IEEE) 5 (5): 8-36, 2017

22 Dimitrios Marmanis, "Deep Learning Earth Observation Classification Using ImageNet Pretrained Networks" Institute of Electrical and Electronics Engineers (IEEE) 13 (13): 105-109, 2016

23 Ke Sun, "Deep High-Resolution Representation Learning for Human Pose Estimation" IEEE 5686-5696, 2019

24 A. L. Westerling, "Climate change and wildfire in California" Springer Science and Business Media LLC 87 (87): 231-249, 2007

25 J. R. Marlon, "Climate and human influences on global biomass burning over the past two millennia" Springer Science and Business Media LLC 1 (1): 697-702, 2008

26 Y. LeCun, "Backpropagation Applied to Handwritten Zip Code Recognition" MIT Press - Journals 1 (1): 541-551, 1989

27 A FERNANDEZ, "Automatic mapping of surfaces affected by forest fires in Spain using AVHRR NDVI composite image data*1" Elsevier BV 60 (60): 153-162, 1997

28 C. Domenikiotis, "Agreement assessment of NOAA/AVHRR NDVI with Landsat TM NDVI for mapping burned forested areas" Informa UK Limited 23 (23): 4235-4246, 2002

29 Nitin Namdeo Pise, "A Survey of Semi-Supervised Learning Methods" IEEE 30-34, 2008

30 Jeff Eidenshink, "A Project for Monitoring Trends in Burn Severity" Springer Science and Business Media LLC 3 (3): 3-21, 2007

31 Korea Forest Service, "2022 Wildfire statistics annual"