基于深度学习的森林可燃物含水率反演技术

doi:10.11707/j.1001-7488.20221005

Abstract

Abstract:

Objective: This paper was carried out to study the water content measurement based on satellite remote sensing data and to compare the accuracy of deep learning model and traditional machine learning model in order to provide a theoretical basis for the establishment of forest fuel water content database in China. Method: Taking Chongli District, Zhangjiakou City, Hebei Province as the research area, based on the field measurement data, aiming at the problem of a large error of traditional machine learning model, the mutilayer perceptron(MLP) model in deep learning was established. The relationships between remote sensing spectral reflectance and water content of forest canopy vegetation and litter were studied, and the accuracy was compared with the support vector regression(SVR) model in traditional machine learning. In the experiment, sentry remote sensing data of the same season as field investigation were selected, and then the model variables commonly used in remote sensing estimation method such as spectral reflectance and spectral moisture index were used as the influence factors of retrieving the moisture content of canopy vegetation and surface litter. Finally, the model training was carried out based on the data of field investigation. Inversion using the remote sensing estimation method for ever encountered in the process of surface litter moisture content of canopy cover problem, this study used in the processing of remote sensing data bidirectional reflectance distribution function to obtain samples of the remote sensing data of different observation angles, combined with the radiative transfer model, the canopy reflectance model mapped to the surface reflectance after training. Result: In the experimental result, the fitting degree of MLP model with red light, green light, near-infrared and short-wave infrared bands as input variables in the inversion of canopy vegetation water content was 0.843, which was better than the fitting degree of the optimal model in SVR of 0.807, and the accuracy improved by 4.5%. The fitting degree of MLP model in the inversion of surface litter water content was 0.448, which was better than the optimal model in SVR of 0.408, and the accuracy improved by 9.8%.In this study, the optimal fitting model MLP was used to invert the distribution map of fuel water content, and it was concluded that the water content of canopy vegetation was higher in the western Chongli district, while the water content of surface litter was higher in the southeastern Chongli district. Conclusion: The research result of this paper could demonstrate an optimization scheme to solve the problem of poor penetrability of optical remote sensing from canopy to surface, and could also provide a theoretical basis for using remote sensing estimation methods to measure moisture content of regional canopy vegetation and surface litter at a large scale.

Key words: remote sensing data, moisture content of forest combustibles, spectral moisture index, deep learning

CLC Number:

S762

Jia Li,Lan Lan,Zuozhong Zhang,Wentao Yuan,Demin Gao,Shuqin Zong,Qiaolin Ye. Inversion Technology of Forest Fuel Moisture Content Based on Deep Learning[J]. Scientia Silvae Sinicae, 2022, 58(10): 47-58.

Figures/Tables 15

Fig.1

Fig.2

Table 1

Fig.3

Fig.4

Fig.5

Fig.6

Fig.7

Table 2

Fig.8

Fig.9

Fig.10

Table 3

Fig.11

Fig.12

References 0

	高超, 林红蕾, 胡海清, 等. 我国林火发生预测模型研究进展. 应用生态学报, 2020, 31 (9): 3227- 3240.
	Gao G , Lin H L , Hu H Q , et al. Research progress of forest fire prediction model in China. Journal of Applied Ecology, 2020, 31 (9): 3227- 3240.
	耿冰, 孙义博, 曾巧林, 等. 基于深度学习方法的PM_(2.5)精细化时空估算模型. 中国环境科学, 2021, 41 (8): 3502- 3510.
	Geng B , Sun Y B , Zeng Q L , et al. PM based on deep learning method_(2.5) refined spatiotemporal estimation model. China Environmental Science, 2021, 41 (8): 3502- 3510.
	郭建军, 张晓峰, 纪月, 等. 崇礼区域生态环境综合评价. 防护林科技, 2021, (1): 10- 12.10-12, 15
	Guo J J , Zhang X F , Ji Y , et al. Comprehensive evaluation of ecological environment in Chongli region. Shelter forest science and technology, 2021, (1): 10- 12.10-12, 15
	郭新彬, 郑文霞, 曾爱聪, 等. 美国林火管理概况及分析. 应用生态学报, 2019, 30 (12): 4361- 4368.
	Guo X B , Zheng W X , Zeng A C , et al. Forest fire management in the United States. Chinese Journal of Applied Ecology, 2019, 30 (12): 4361- 4368.
	刘曦, 金森. 平衡含水率法预测死可燃物含水率的研究进展. 林业科学, 2007, 43 (12): 126- 133.
	Liu X , Jin S . research progress in predicting moisture content of dead combustibles by equilibrium moisture content method. Forestry Science, 2007, 43 (12): 126- 133.
	刘昕, 邸雪颖. 三种方法对森林地表可燃物含水率的预测评价. 森林工程, 2013, 29 (2): 8- 13.8-13, 20
	Liu X , Di X Y . Prediction and evaluation of moisture content of forest surface combustibles by three methods. Forest engineering, 2013, 29 (2): 8- 13.8-13, 20
	吕杰. 2012. 基于机器学习和辐射传输模型的农作物叶绿素含量高光谱反演模型. 北京: 中国地质大学(北京).
	Lü J. 2012. Hyperspectral inversion model of crop chlorophyll content based on machine learning and radiative transfer model. Beijing: China University of Geosciences(Beijing). [in Chinese]
	潘佩芬, 杨武年, 简季, 等. 基于光谱指数的植被含水率遥感反演模型研究——以岷江上游毛尔盖地区为例. 遥感信息, 2013, 28 (3): 69- 73.
	Pan P F , Yang W N , Jian J , et al. Study on remote sensing inversion model of vegetation moisture content based on spectral index—Taking Maoergai area in the upper reaches of Minjiang River as an example. Remote Sensing Information, 2013, 28 (3): 69- 73.
	申洪源, 马亮, 张雅楠, 等. 基于多传感器和SVR算法的油田多相流实时计量技术研究. 仪器仪表用户, 2019, 26 (10): 15- 19.
	Shen H Y , Ma L , Zhang Y N , et al. Research on real-time measurement technology of oilfield multiphase flow based on multisensor and SVR algorithm. Instrument Users, 2019, 26 (10): 15- 19.
	王熊. 2020. 基于哨兵二号的森林动态变化检测方法研究. 武汉: 华中农业大学.
	Wang X. 2020. Study on forest dynamic change detection method based on sentry 2. Wuhan: Huazhong Agricultural University. [in Chinese]
	谢字希. 2016. 黑龙江大兴安岭地区森林枯落物含水率遥感反演. 哈尔滨: 东北林业大学.
	Xie Z X. 2016. Remote sensing inversion of forest litter moisture content in Daxing'anling area, Heilongjiang Province. Harbin: Northeast Forestry University. [in Chinese]
	谢字希, 胡海清, 杨曦光, 等. 基于实测光谱的大兴安岭地区典型森林枯落物含水率估测模型. 生态学杂志, 2017, 36 (11): 3321- 3328.
	Xie Z X , Hu H Q , Yang X G , et al. Estimation model of water content of typical forest litter in Daxing'anling area based on measured spectrum. Journal of Ecology, 2017, 36 (11): 3321- 3328.
	张佳华, 许云, 姚凤梅, 等. 植被含水量光学遥感估算方法研究进展. 中国科学: 技术科学, 2010, 40 (10): 1121- 1129.
	Zhang J H , Xu Y , Yao F M , et al. Research progress on optical remote sensing estimation method of vegetation water content. Chinese Science: Technical Science, 2010, 40 (10): 1121- 1129.
	张恒, 刘鑫源, 李星汉, 等. 我国气象要素回归法预测地表可燃物含水率研究进展. 森林防火, 2018, (3): 25- 31.
	Zhang H , Liu X Y , Li X H , et al. Research progress on prediction of surface combustible moisture content by meteorological factor regression method in China. Forest Fire Prevention, 2018, (3): 25- 31.
	赵美霞. 2021. 北京2022年冬奥会背景下河北省冰雪运动政策研究. 北京: 首都体育学院.
	Zhao M X. 2021. Research on ice and snow sports policy of Hebei Province under the background of Beijing 2022 Winter Olympic Games. Beijing: Capital Institute of Physical Education. [in Chinese]
	赵永光, 马灵玲, 李传荣, 等. 基于冠层辐射传输模型的地表反射率光谱重建方法. 光谱学与光谱分析, 2015, 35 (7): 1763- 1769.
	Zhao Y G , Ma L L , Li C R , et al. Spectral reconstruction method of surface reflectance based on canopy radiative transfer model. Spectroscopy and Spectral Analysis, 2015, 35 (7): 1763- 1769.
	Barmpoutis P , Papaioannou P , Dimitropoulos K , et al. A review on early forest fire detection systems using optical remote sensing. Sensors, 2020, 20 (22)
	Maffei C , Lindenbergh R , Menenti M . Combining multi-spectral and thermal remote sensing to predict forest fire characteristics. ISPRS Journal of Photogrammetry and Remote Sensing, 2021, 181, 400- 412.
	Mohajane M , Costache R , Karimi F , et al. Application of remote sensing and machine learning algorithms for forest fire mapping in a Mediterranean area. Ecological Indicators, 2021, 129, 107869.
	Moritz M A , Batllori E , Bradstock R A . Learning to coexist with wildfire. Nature, 2014, 515, 58- 66.
	Pisek J , Erb A , Korhonen L , et al. Retrieval and validation of forest background reflectivity from daily Moderate Resolution Imaging Spectroradiometer(MODIS) bidirectional reflectance distribution function(BRDF) data across European forests. Biogeosciences, 2021, 18 (2): 621- 635.
	Quan X , Yebra M , Riaño D , et al. Global fuel moisture content mapping from MODIS. International Journal of Applied Earth Observations and Geoinformation, 2021, 101, 102354.
	Schneider F D , l Kükenbrink D , Schaepman M E , et al. Quantifying 3D structure and occlusion in dense tropical and temperate forests using close-range LiDAR. Agricultural and Forest Meteorology, 2019, 268
	Ye Q , Huang P , Zhang Z , et al. Multiview learning with robust double-sided twin SVM. IEEE Transactions on Cybernetics, 2021,

水分指数 Water index	光谱指数 Spectral indices	计算公式 Formula
比值水分指数 Specific moisture index	简单比 Simple ratio，SR	$\frac{\rho_{895}}{\rho_{675}}$
比值水分指数 Specific moisture index	水分指数 Water index, WI	$\frac{\rho_{900}}{\rho_{970}}$
比值水分指数 Specific moisture index	水分胁迫指数 Water stress index, WSI	$\frac{\rho_{1600}}{\rho_{820}}$
比值水分指数 Specific moisture index	简单水分指数 Simple ratio of water index，SRWI	$\frac{\rho_{860}}{\rho_{1240}}$
归一化水分指数 Normalized water index	归一化植被指数 Normalized difference vegetation index，NDVI	$\frac{\rho_{895}-\rho_{675}}{\rho_{895}+\rho_{675}}$
归一化水分指数 Normalized water index	归一化差异水分指数 Normalized difference water index，NDWI	$\frac{\rho_{860}-\rho_{1240}}{\rho_{860}+\rho_{1240}}$
归一化水分指数 Normalized water index	归一化红外指数 Normalized difference infrared index，NDII	$\frac{\rho_{820}-\rho_{1600}}{\rho_{820}+\rho_{1600}}$
复比值型指数 Complex ratio index	WI/NDVI	$\frac{\rho_{900} / \rho_{970}}{\frac{\rho_{895}-\rho_{675}}{\rho_{895}+\rho_{675}}}$

自变量 Independent variables	拟合模型 Fitting model	拟合度 Degree of fitting(R²)
原始波段Original band	MLP	0.843
SRWI	SVR	0.644
NDII	SVR	0.807
NDVI	SVR	0.703
NDWI	SVR	0.678

自变量 Independent variables	拟合模型 Fitting model	拟合度 Degree of fitting(R²)
原始波段Original band	MLP	0.448
SR	SVR	0.367
NDII	SVR	0.360
NDVI	SVR	0.408
NDWI	SVR	0.361

[1]	Ying Pan,Mingming Ding,Jie Lin,Qiao Dai,Geng Guo,Linlin Cui. Inversion of Forest Leaf Area Index Based on PROSAIL Model and Multi-Angle Remote Sensing Data [J]. Scientia Silvae Sinicae, 2021, 57(4): 90-106.
[2]	Yan Zhou,Wenping Liu,Youqing Luo,Shixiang Zong. Small Object Detection for Infected Trees Based on the Deep Learning Method [J]. Scientia Silvae Sinicae, 2021, 57(3): 98-107.
[3]	Ying Guo,Zengyuan Li,Erxue Chen,Xu Zhang,Lei Zhao,Yan Chen,Yahui Wang. A Deep Learning Method for Forest Fine Classification Based on High Resolution Remote Sensing Images: Two-Branch FCN-8s [J]. Scientia Silvae Sinicae, 2020, 56(3): 48-60.
[4]	Mingwei Shao,Junyu Dong. A New Algorithm for Automatic Optimizing Cross-Cut Saw Based on Deep Learning Algorithm [J]. Scientia Silvae Sinicae, 2020, 56(12): 123-129.
[5]	Liu Luxia, Pang Yong, Ren Haibao, Li Zengyuan. Predict Tree Species Diversity from GF-2 Satellite Data in a Subtropical Forest of China [J]. Scientia Silvae Sinicae, 2019, 55(2): 61-74.
[6]	Qiu Ni, Xu Songjun, Qiu Penghua, Yang Wenhuai, Yang Xiu, Yang Qian. Community Distribution and Landscape Pattern of the Mangrove on the Qi'ao Island, Zhuhai [J]. Scientia Silvae Sinicae, 2019, 55(1): 1-10.
[7]	Li Xiaohong, Chen Erxue, Li Zengyuan, Li Shiming. Object Based Land Cover Classification Method Integrating Multi-Source Remote Sensing Data [J]. Scientia Silvae Sinicae, 2018, 54(2): 68-80.
[8]	Zu Xiaofeng, Qin Xianlin, Li Zengyuan, Sun Guifen, Liu Shuchao. Method for Burned Forest Biomass Estimation Using Satellite Remote Sensing Based on Power Law Distribution [J]. Scientia Silvae Sinicae, 2017, 53(10): 90-99.
[9]	Zheng Wei;Li Yajun;Liu Cheng;Wang Meng. Extracting Forest Burned Scar Region Area Based on Multi-Source Remote Sensing Data [J]. Scientia Silvae Sinicae, 2011, 47(8): 192-195.
[10]	Tan Bingxiang;Du Jishan. Analysis of Vegetation and Soil Erosion for Forest Areas Using Remote Sensing Data [J]. Scientia Silvae Sinicae, 2006, 42(4): 7-11.

Inversion Technology of Forest Fuel Moisture Content Based on Deep Learning

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 15

References 0

Related Articles 10

Recommended Articles

Metrics

Comments