南方丘陵区林下植被覆盖度无人机多角度遥感测量

doi:10.16258/j.cnki.1674-5906.2021.12.003

生态环境学报 ›› 2021, Vol. 30 ›› Issue (12): 2294-2302.DOI: 10.16258/j.cnki.1674-5906.2021.12.003

南方丘陵区林下植被覆盖度无人机多角度遥感测量

王瑞璠¹^,²(), 魏倪彬¹^,², 张仓皓¹^,², 鲍甜甜¹^,², 刘健¹^,², 余坤勇¹^,², 王帆¹^,²^,^*()

1.福建农林大学林学院,福建福州 350002
2.福建农林大学/3s技术与资源优化利用福建省高等学校重点实验室,福建福州 350002

收稿日期:2020-07-05 出版日期:2021-12-18 发布日期:2022-01-04
通讯作者: *王帆（1987年生）,男,讲师,研究方向为水土流失遥感监测。E-mail: pipi870408@163.com
作者简介:王瑞璠（1998年生）,男,硕士研究生,研究方向为植被参数遥感量化。E-mail: 1264114710@qq.com
基金资助:
国家青年科学基金项目(41901387);福建省教育厅省中青年教师教育科研项目(JT180132)

UAV Multi Angle Remote Sensing Quantification of Understory Vegetation Coverage in the Hilly Region of South China

WANG Ruifan¹^,²(), WEI Nibin¹^,², ZHANG Canghao¹^,², BAO Tiantian¹^,², LIU Jian¹^,², YU Kunyong¹^,², WANG Fan¹^,²^,^*()

1. College of Forestry, Fujian Agriculture and Forestry University, Fuzhou 350002, China
2. 3S Technology and Resources Optimized Utilization Key Laboratory of Fujian University, Fuzhou 350002, China

Received:2020-07-05 Online:2021-12-18 Published:2022-01-04

摘要/Abstract

摘要：

林下植被覆盖度作为描述森林生态系统健康的重要参量,是探查水土流失情况的重要指征。量化南方丘陵地带地形起伏区林下植被覆盖度对水土流失精准治理具有重要意义。选取长汀县河田镇作为研究区,基于无人机多角度（0°、10°、20°、30°）遥感数据,利用局部最大值法,将植被形态划分为离散型与连续型,并分别利用点云布料滤波与地图森林密度算法反演出对应植被形态的林下地形。采用半高斯拟合（HAGFVC）法分离植被,再设置高程差阈值提取出林下植被,结合不同角度影像,利用形态学中的斑块膨胀方法弥补树冠遮挡的误差,进而量化林下植被覆盖度。结果表明,针对不同植被形态,上述技术路线所得的地形反演结果精度较高,其中离散型样地RMSE为0.400 m,连续型样地RMSE为0.518 m。单一角度中,基于20°影像估算林下植被覆盖度的结果精度最高（R²=0.459,RMSE=0.166）;基于30°影像的估算精度最低（R²=0.337,RMSE=0.243）。多角度影像耦合（0°+10°+20°+30°）后,林下植被覆盖度的估算精度得到明显提高,R²达到0.675,RMSE为0.102。研究结果可以为南方红壤区林下植被信息无人机遥感调查提供技术支撑。

关键词: 南方丘陵区, 林下植被覆盖度, 地形反演, 无人机, 多角度遥感

Abstract:

As an important parameter to describe the forest ecosystem health, understory vegetation coverage is an important indicator to explore the situation of soil and water loss. Quantifying the understory vegetation coverage in the hilly area of south China is of great significance to the accurate control of soil and water loss. In this paper, Hetian Town, Changting County was selected as the study area. Based on the UAV multi angle (0°, 10°, 20°, 30°) remote sensing data, the vegetation form was divided into discrete and continuous types by using the local maximum method, and the understory terrain corresponding to the vegetation form was inversed by using the point cloud distribution filter and the map forest density algorithm respectively. The semi Gaussian fitting (HAGFVC) method was used to separate the soil background and vegetation information, and the elevation difference threshold was set to extract the understory vegetation. Combined with different angle images, the patch expansion method in morphology was used to make up for the error of canopy occlusion, and then quantified the understory vegetation coverage. The results showed that for both of vegetation types, the terrain inversion results have high accuracy, the RMSE of discrete sample plot was 0.400 m and that of continuous sample plot was 0.518 m. For single angle, the estimation accuracy based on the 20° image was the highest (R²=0.459, RMSE=0.166); the estimation accuracy based on 30° image was the lowest (R²=0.337, RMSE=0.243). With multi angle images coupled (0°+10°+20°+30°), the estimation accuracy was significantly improved, R² reached 0.675 and RMSE was 0.102. The results of this paper could provide technical support for UAV remote sensing investigation of understory vegetation information in the red soil area of South China.

Key words: southern hilly area, understory vegetation cover, terrain inversion, unmanned aerial vehicle, multi angle remote sensing

中图分类号:

S771.8
X173

王瑞璠, 魏倪彬, 张仓皓, 鲍甜甜, 刘健, 余坤勇, 王帆. 南方丘陵区林下植被覆盖度无人机多角度遥感测量[J]. 生态环境学报, 2021, 30(12): 2294-2302.

WANG Ruifan, WEI Nibin, ZHANG Canghao, BAO Tiantian, LIU Jian, YU Kunyong, WANG Fan. UAV Multi Angle Remote Sensing Quantification of Understory Vegetation Coverage in the Hilly Region of South China[J]. Ecology and Environment, 2021, 30(12): 2294-2302.

图/表 9

参考文献 32

[1]	ACHANTA R, SHAJI A, SMITH K, et al., 2012. SLIC superpixels compared to state-of-the-art superpixel methods[J]. IEEE transactions on pattern analysis and machine intelligence, 34(11): 2274-2282. DOI URL
[2]	ANDERSON C T, DIETZ S L, POKSWINSKI S M, et al., 2021. Traditional field metrics and terrestrial LiDAR predict plant richness in southern pine forests[J]. Forest Ecology and Management, DOI: 10.1016/j.foreco.2021.119118. DOI
[3]	CHEN Y Q, ZHANG Y, YU S Q, et al., 2021. Responses of soil labile organic carbon and water-stable aggregates to reforestation in southern subtropical China[J]. Journal of Plant Ecology, 14(2): 191-201. DOI URL
[4]	FORREST G H, THOMAS H, NICHOLAS C C, et al., 2008. Multi-angle remote sensing of forest light use efficiency by observing PRI variation with canopy shadow fraction[J]. Remote Sensing of Environment, 112(7): 3201-3211. DOI URL
[5]	HARTLEY R J L, LEONARDO E M, MASSAM P et al., 2020. An Assessment of High-Density UAV Point Clouds for the Measurement of Young Forestry Trials[J]. Remote Sensing, DOI: 10.3390/rs12244039. DOI
[6]	KLÁPŠTĚ P, FOGL M, BARTÁK V, et al., 2020. Sensitivity analysis of parameters and contrasting performance of ground filtering algorithms with UAV photogrammetry-based and LiDAR point clouds[J]. International Journal of Digital Earth, 13(12): 1672-1694. DOI URL
[7]	LIN L C, YU K Y, YAO X, et al., 2021. UAV Based Estimation of Forest Leaf Area Index (LAI) through Oblique Photogrammetry[J]. Remote Sensing, 13(4): 803. DOI URL
[8]	LI L Y, CHEN J, MU X H, et al., 2020. Quantifying Understory and Overstory Vegetation Cover Using UAV-Based RGB Imagery in Forest Plantation[J]. Remote Sensing, 12(2): 298. DOI URL
[9]	LI L Y, MU X H, MACFARLANE C, et al., 2018. A half-Gaussian fitting method for estimating fractional vegetation cover of corn crops using unmanned aerial vehicle images[J]. Agricultural and Forest Meteorology, 262: 379-390. DOI URL
[10]	MAALEK R, 2021. Field information modeling (FIM)^TM: Best practices using point clouds[J]. Remote Sensing, DOI: 10.3390/rs13050967. DOI
[11]	NA J M, XUE K K, XIONG L Y, et al., 2020. UAV-based terrain modeling under vegetation in the Chinese Loess Plateau: A deep learning and terrain correction ensemble framework[J]. Remote Sensing, DOI: 10.3390/rs12203318. DOI
[12]	NIEDERHEISER R, WINKLER M, DI CECCO V, et al., 2021. Using automated vegetation cover estimation from close-range photogrammetric point clouds to compare vegetation location properties in mountain terrain[J]. GIScience & Remote Sensing, 58(1): 120-137.
[13]	PANDEY S, KUMAR P, ZLATIC M, et al., 2021. Recent advances in assessment of soil erosion vulnerability in a watershed[J]. International Soil and Water Conservation Research, 9(3): 305-318. DOI URL
[14]	SHARDA V N, MANDAL D, DOGRA P, 2021. Prioritizing soil conservation measures based on water erosion risk and production and bio-energy losses in peninsular South Indian states[J]. Catena, DOI: 10.1016/j.catena.2021.105263. DOI
[15]	TOLGA G, 2019. Landslide recognition and mapping in a mixed forest environment from airborne LiDAR data[J]. Engineering Geology, DOI: 10.1016/j.enggeo.2019.105155. DOI
[16]	ZHANG W M, QI J B, PENG W, et al., 2016. An Easy-to-Use Airborne LiDAR Data Filtering Method Based on Cloth Simulation[J]. Remote Sensing, DOI: 10.3390/rs8060501. DOI
[17]	MENG X L, SHANG N, ZHANG X K, et al., 2017. Photogrammetric UAV Mapping of Terrain under Dense Coastal Vegetation: An Object-Oriented Classification Ensemble Algorithm for Classification and Terrain Correction[J]. Multidisciplinary Digital Publishing Institute, DOI: 10.3390/rs9111187. DOI
[18]	YAN Y J, DENG H Z, LIU Y et al., 2019. Application of UAV-based multi-angle hyperspectral remote sensing in fine vegetation classification[J]. Remote Sensing, 11(23): 2753. DOI URL
[19]	YANG L, RONGGAO L, JAN P, et al., 2017. Separating overstory and understory leaf area indices for global needleleaf and deciduous broadleaf forests by fusion of MODIS and MISR data[J]. Biogeosciences, 14(5): 1093-1110. DOI URL
[20]	YAO Y, LEUNG Y, FUNG T, et al., 2021. Continuous multi-angle remote sensing and its application in urban land cover classification[J]. Remote Sensing, DOI: 10.3390/rs13030413. DOI
[21]	YOLANDA S S, ANTONIO M, FERNANDO S F, et al., 2018. Mapping Wildfire Ignition Probability Using Sentinel 2 and LiDAR Jerte Valley, Cáceres, Spain[J]. Sensors, 18(3): 826. DOI URL
[22]	陈秋计, 杭梦如, 李继业, 等, 2020. 基于LIDAR的矿区恢复植被高度提取方法研究[J]. 煤炭科学技术, 48(4): 113-119.
	CHEN Q J, HANG M R, LI J Y, et al., 2020. Extraction method of vegetation height in mining area based on lidar[J]. Coal science and technology, 48(4): 113-119.
[23]	韩文霆, 汤建栋, 张立元, 等, 2021. 基于无人机遥感玉米水分利用效率及生物量监测[J]. 农业机械学报, 52(5): 129-141.
	HAN W T, TANG J D, ZHANG L Y, et al., 2021. Monitoring of water use efficiency and biomass of Maize Based on UAV Remote Sensing[J]. Transactions of the Chinese Society for Agricultural Machinery, 52(5): 129-141.
[24]	胡小芳, 李小雅, 赵红敏, 等, 2020. 民宿价格的空间分异特征及影响因素--以湖北省恩施州为例[J]. 自然资源学报, 35(10): 2473-2483.
	HU X F, LI X Y, ZHAO H M, et al., 2020. Spatial variation characteristics and influencing factors of homestay inn price: A case study of Enshi, Hubei[J]. Journal of natural resources, 35(10): 2473-2483. DOI URL
[25]	江海英, 贾坤, 赵祥, 等, 2020. 山地叶面积指数反演理论、方法与研究进展[J]. 遥感学报, 24(12): 1433-1449.
	JIANG H Y, JIA K, ZHAO X, et al., 2020. Inversion theory, method and research progress of mountain leaf area index[J]. National Remote Sensing Bulletin, 24(12): 1433-1449.
[26]	焦桐, 刘荣高, 刘洋, 等, 2014. 林下植被遥感反演研究进展[J]. 地球信息科学学报, 16(4): 602-608. DOI
	JIAO T, LIU R G, LIU Y, et al., 2014. Advances in remote sensing inversion of understory vegetation[J]. Journal of Geo-information Science, 16(4): 602-608.
[27]	王敬哲, 陈志强, 陈志彪, 等, 2020. 南方红壤侵蚀区不同植被恢复年限下芒萁叶功能性状对土壤因子的响应[J]. 生态学报, 40(3): 900-909.
	WANG J Z, CHEN Z Q, CHEN Z B, et al., 2020. Response of leaf functional traits of Osmunda japonica to soil factors under different vegetation restoration years in red soil erosion area of southern China[J]. Acta ecologica Sinica, 40(3): 900-909.
[28]	徐逸, 甄佳宁, 蒋侠朋, 等, 2021. 无人机遥感与XGBoost的红树林物种分类[J]. 遥感学报, 25(3): 737-752.
	XU Y, ZHEN J N, JIANG X P, et al., 2021. Classification of mangrove species based on UAV remote sensing and xgboost[J]. National Remote Sensing Bulletin, 25(3): 737-752.
[29]	阎广建, 姜海兰, 闫凯, 等, 2021. 多角度光学定量遥感[J]. 遥感学报, 25(1): 83-108.
	YAN G J, JIANG H L, YAN K, et al., 2021. Multi angle optical quantitative remote sensing[J]. National Remote Sensing Bulletin, 25(1): 83-108.
[30]	杨蜀秦, 宋志双, 尹瀚平, 等, 2021. 基于深度语义分割的无人机多光谱遥感作物分类方法[J]. 农业机械学报, 52(3): 185-192.
	YANG S Q, SONG Z S, YIN H P, et al., 2021. Crop classification method based on depth semantic segmentation for UAV multispectral remote sensing[J]. Transactions of the Chinese Society for Agricultural Machinery, 52(3): 185-192.
[31]	张仓皓, 杨樟平, 谢巧雅, 等, 2020. 毛竹立竹度无人机遥感识别有效高度的研究[J]. 遥感技术与应用, 35(6): 1436-1446.
	ZHANG C H, YANG Z P, XIE Q Y, et al., 2020. Study on effective height recognition of Phyllostachys pubescens based on UAV Remote Sensing[J]. Remote Sensing Technology and Application, 35(6): 1436-1446.
[32]	赵长森, 潘旭, 杨胜天, 等, 2019. 低空遥感无人机影像反演河道流量[J]. 地理学报, 74(7): 1392-1408. DOI
	ZHAO C S, PAN X, YANG S T, et al., 2019. Low altitude remote sensing UAV image retrieval of river flow[J]. Acta Geographica Sinica, 74(7): 1392-1408.

植被形态 Vegetation morphology	树冠点提取结果 Canopy point extraction results	地形反演结果 Terrain inversion results
离散 Discrete
连续 Continuous

植被形态 Vegetation morphology	树冠点提取结果 Canopy point extraction results	地形反演结果 Terrain inversion results
离散 Discrete
连续 Continuous

样地编号 Plot number	植被形态 Vegetation morphology	郁闭度 Canopy closure	平均坡度 Average Slope/(°)	地形起伏 Undulating Terrain/m	RMSE_cloth/ m	RMSE_map/ m
1	离散 Discrete	0.081	9.466	4.947	0.303	0.584
2	离散 Discrete	0.218	19.053	9.067	0.069	0.412
3	离散 Discrete	0.323	22.268	10.06	0.067	0.355
4	离散 Discrete	0.156	17.596	8.364	0.898	1.856
5	离散 Discrete	0.248	22.404	10.065	0.107	0.995
6	连续 Continuous	0.352	22.567	9.688	3.356	0.757
7	连续 Continuous	0.27	15.779	5.17	1.547	0.210
8	连续 Continuous	0.148	22.781	9.525	0.985	0.194
9	连续 Continuous	0.482	5.341	2.606	3.528	0.918
10	离散 Discrete	0.168	9.359	6.667	0.761	1.133
11	连续 Continuous	0.482	18.676	15.014	2.857	0.681
12	离散 Discrete	0.302	17.617	8.163	0.775	1.756
13	离散 Discrete	0.178	19.872	9.134	0.222	0.574
14	连续 Continuous	0.445	5.844	2.199	1.894	0.348

样地编号 Plot number	植被形态 Vegetation morphology	郁闭度 Canopy closure	平均坡度 Average Slope/(°)	地形起伏 Undulating Terrain/m	RMSE_cloth/ m	RMSE_map/ m
1	离散 Discrete	0.081	9.466	4.947	0.303	0.584
2	离散 Discrete	0.218	19.053	9.067	0.069	0.412
3	离散 Discrete	0.323	22.268	10.06	0.067	0.355
4	离散 Discrete	0.156	17.596	8.364	0.898	1.856
5	离散 Discrete	0.248	22.404	10.065	0.107	0.995
6	连续 Continuous	0.352	22.567	9.688	3.356	0.757
7	连续 Continuous	0.27	15.779	5.17	1.547	0.210
8	连续 Continuous	0.148	22.781	9.525	0.985	0.194
9	连续 Continuous	0.482	5.341	2.606	3.528	0.918
10	离散 Discrete	0.168	9.359	6.667	0.761	1.133
11	连续 Continuous	0.482	18.676	15.014	2.857	0.681
12	离散 Discrete	0.302	17.617	8.163	0.775	1.756
13	离散 Discrete	0.178	19.872	9.134	0.222	0.574
14	连续 Continuous	0.445	5.844	2.199	1.894	0.348

南方丘陵区林下植被覆盖度无人机多角度遥感测量

UAV Multi Angle Remote Sensing Quantification of Understory Vegetation Coverage in the Hilly Region of South China

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 32

相关文章 0

编辑推荐

Metrics

本文评价