Aboveground biomass estimation of Zhaosu mountain meadow based on visible light images with different resolution

doi:10.6048/j.issn.1001-4330.2025.01.027

Abstract

Abstract:

【Objective】 Unmanned aerial vehicles (UAVs) equipped with digital cameras offer high mobility, flexibility, and resolution, making them advantageous for rapid and accurate AGB estimation. This study aims to investigate the impact of image resolution differences at various flight altitudes of UAVs on the accuracy of AGB estimation. 【Methods】 In this study, we conducted UAV flights at five different altitudes (10, 30, 50, 70, 90 m) over the Zhaosu mountain meadow in Xinjiang to explore the effects of image acquisition at different flight altitudes on AGB estimation accuracy by analyzing differences in spectral information and texture features. 【Results】 By extracting digital images collected at different flight altitudes, we analyzed their spectral information and texture features and correlated these features with measured AGB.The top 8 vegetation indices selected in this study and the top 8 texture features showed strong correlations with AGB. After integrating the three input variables’ variance inflation factor (VIF), we constructed an AGB estimation model using Principal Component Analysis (PCA) and Multiple Linear Regression (MLR). The study compared the model accuracy for estimating AGB using images of different resolutions. 【Conclusion】 The results indicate that: (1) The correlation between texture features and AGB is weaker than that of vegetation indices when image resolution ranges from 0.27 to 2.45 cm but both of them reach a significant level. As image resolution decreases, the difference in correlation between the two becomes more apparent. (2) At the same image resolution, the best AGB estimation results are achieved when spectral information is combined with texture features, followed by a model using a single texture feature, with a single spectral model performing the worst. (3) As image resolution increases, the accuracy of AGB estimation improves for models using spectral information, texture information, and spectral + texture information.

Key words: grassland; above-ground biomass; image resolution; spectral information; texture features

摘要：

【目的】 无人机搭载数码相机具有较高的机动性、灵活性和分辨能力,在快速、准确估算草地AGB等方面优势明显。【方法】 探讨无人机搭载的数码相机在不同飞行高度的影像分辨率差对草地AGB估算精度,在新疆昭苏山地草甸设置10、30、50、70和90 m 5个飞行高度,探究不同飞行高度下获取的影像在光谱信息、纹理特征等差异下对草地AGB估算精度的影响。【结果】 在相同的飞行高度下,采用光谱信息与纹理特征相结合的方法相较于单独使用光谱信息或纹理特征,可以提高AGB估算的精度,分别提高了22.34%、13.25%、11.11%、2.18%和2.35%。仅依赖数码图像的光谱信息来估算草地AGB容易导致饱和现象。然而,与光谱信息相比,草地的纹理特征受环境影响较小,在相同的图像分辨率下,所获得的模型效果更佳,弥补单一指标估算草地AGB的不足。【结论】 影像分辨率在0.27~2.45 cm时,纹理特征与草地AGB的相关性弱于植被指数,但均达到显著水平,随着图像清晰度减低,两者之间的关联性差异变得显著;在同种图像分辨率前提下,将光谱信息与纹理特征相结合可以实现最佳的草地AGB估算效果,单一的纹理特征模型次之,单一的光谱模型效果最差;随着图像分辨率的递增,对草地AGB的估算精度受到光谱信息、纹理信息以及光谱与纹理信息的影响呈现上升趋势。

关键词: 草地, 地上生物量, 影像分辨率, 光谱信息, 纹理特征

CLC Number:

S812

YUAN Yilin, YAN An, NING Songrui, HOU Zhengqing, ZHANG Zhenfei, XIAO Shuting, SUN Zhe, XIA Wenqiu. Aboveground biomass estimation of Zhaosu mountain meadow based on visible light images with different resolution[J]. Xinjiang Agricultural Sciences, 2025, 62(1): 234-242.

袁以琳, 颜安, 宁松瑞, 侯正清, 张振飞, 肖淑婷, 孙哲, 夏雯秋. 基于可见光不同分辨率影像下的昭苏山地草甸地上生物量估算[J]. 新疆农业科学, 2025, 62(1): 234-242.

Figures/Tables 7

References 22

[1]	刘伟, 周华坤, 周立. 不同程度退化草地生物量的分布模式[J]. 中国草地, 2005, 27(2): 9-15.
	LIU Wei, ZHOU Huakun, ZHOU Li. Biomass distribution pattern of degraded grassland in alpine meadow[J]. Grassland of China, 2005, 27(2): 9-15.
[2]	赖炽敏, 赖日文, 薛娴, 等. 基于植被盖度和高度的不同退化程度高寒草地地上生物量估算[J]. 中国沙漠, 2019, 39(5): 127-134. DOI
	LAI Chimin, LAI Riwen, XUE Xian, et al. Estimation of aboveground biomass of different degraded alpine grassland based on vegetation coverage and height[J]. Journal of Desert Research, 2019, 39(5): 127-134. DOI
[3]	李宗南, 陈仲新, 王利民, 等. 基于小型无人机遥感的玉米倒伏面积提取[J]. 农业工程学报, 2014, 30(19): 207-213.
	LI Zongnan, CHEN Zhongxin, WANG Limin, et al. Area extraction of maize lodging based on remote sensing by small unmanned aerial vehicle[J]. Transactions of the Chinese Society of Agricultural Engineering, 2014, 30(19): 207-213.
[4]	陈鹏飞, 梁飞. 基于低空无人机影像光谱和纹理特征的棉花氮素营养诊断研究[J]. 中国农业科学, 2019, 52(13): 2220-2229. DOI
	CHEN Pengfei, LIANG Fei. Cotton nitrogen nutrition diagnosis based on spectrum and texture feature of images from low altitude unmanned aerial vehicle[J]. Scientia Agricultura Sinica, 2019, 52(13): 2220-2229. DOI
[5]	干友民, 成平, 周纯兵, 等. 若尔盖亚高山草甸地上生物量与植被指数关系研究[J]. 自然资源学报, 2009, 24(11): 1963-1972. DOI
	GAN Youmin, CHENG Ping, ZHOU Chunbing, et al. Study on pertinence between the vegetation indexes and the aboveground biomass of sub-alpine meadow of zoige county[J]. Journal of Natural Resources, 2009, 24(11): 1963-1972. DOI
[6]	李淑贞, 徐大伟, 范凯凯, 等. 基于无人机与卫星遥感的草原地上生物量反演研究[J]. 遥感技术与应用, 2022, 37(1): 272-278. DOI
	LI Shuzhen, XU Dawei, FAN Kaikai, et al. Research of grassland aboveground biomass inversion based on UAV and satellite remoting sensing[J]. Remote Sensing Technology and Application, 2022, 37(1): 272-278.
[7]	张正健, 李爱农, 边金虎, 等. 基于无人机影像可见光植被指数的若尔盖草地地上生物量估算研究[J]. 遥感技术与应用, 2016, 31(1): 51-62.
	ZHANG Zhengjian, LI Ainong, BIAN Jinhu, et al. Estimating aboveground biomass of grassland in zoige by visible vegetation index derived from unmanned aerial vehicle image[J]. Remote Sensing Technology and Application, 2016, 31(1): 51-62.
[8]	杨琦, 叶豪, 黄凯, 等. 利用无人机影像构建作物表面模型估测甘蔗LAI[J]. 农业工程学报, 2017, 33(8): 104-111.
	YANG Qi, YE Hao, HUANG Kai, et al. Estimation of leaf area index of sugarcane using crop surface model based on UAV image[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(8): 104-111.
[9]	Yue J B, Feng H K, Jin X L, et al. A comparison of crop parameters estimation using images from UAV-mounted snapshot hyperspectral sensor and high-definition digital camera[J]. Remote Sensing, 2018, 10(7): 1138.
[10]	卜灵心, 来全, 刘心怡. 不同机器学习算法在草原草地生物量估算上的适应性研究[J]. 草地学报, 2022, 30(11): 3156-3164. DOI
	BU Lingxin, LAI Quan, LIU Xinyi. Adaptation of different machine learning algorithms for grassland biomass estimation[J]. Acta Agrestia Sinica, 2022, 30(11): 3156-3164. DOI
[11]	刘武超, 罗研, 马瞳宇, 等. 高分辨率遥感影像多特征灌丛提取方法研究[J]. 测绘与空间地理信息, 2022, 45(S1): 111-113.
	LIU Wuchao, LUO Yan, MA Tongyu, et al. Study on extraction method of multi-feature shrub from high-resolution remote sensing image[J]. Geomatics & Spatial Information Technology, 2022, 45(S1): 111-113.
[12]	李健, 江洪, 罗文彬, 等. 融合无人机多光谱和纹理特征的马铃薯LAI估算[J]. 华南农业大学学报, 2023, 44(1): 93-101.
	LI Jian, JIANG Hong, LUO Wenbin, et al. Potato LAI estimation by fusing UAV multi-spectral and texture features[J]. Journal of South China Agricultural University, 2023, 44(1): 93-101.
[13]	杨福芹, 冯海宽, 肖天豪, 等. 融合无人机影像光谱与纹理特征的冬小麦氮营养指数估算[J]. 农业现代化研究, 2020, 41(4): 718-726.
	YANG Fuqin, FENG Haikuan, XIAO Tianhao, et al. Nitrogen nutrition index estimation in winter wheat by UAV spectral information and texture feature fusion[J]. Research of Agricultural Modernization, 2020, 41(4): 718-726.
[14]	刘畅, 杨贵军, 李振海, 等. 融合无人机光谱信息与纹理信息的冬小麦生物量估测[J]. 中国农业科学, 2018, 51(16): 3060-3073. DOI
	LIU Chang, YANG Guijun, LI Zhenhai, et al. Biomass estimation in winter wheat by UAV spectral information and texture information fusion[J]. Scientia Agricultura Sinica, 2018, 51(16): 3060-3073. DOI
[15]	朱永基, 殷飞笺, 王晗, 等. 随机森林方法支持下Sentinel-2A MSI多种特征的冬小麦识别分析[J]. 安徽科技学院学报, 2022, 36(1): 25-31.
	ZHU Yongji, YIN Feijian, WANG Han, et al. Winter wheat identification based on multiple features of sentinel-2AMSI data and random forest method[J]. Journal of Anhui Science and Technology University, 2022, 36(1): 25-31.
[16]	黎松松, 王宁欣, 郑伟, 等. 一年生和多年生豆禾混播草地超产与多样性效应的比较[J]. 植物生态学报, 2021, 45(1): 23-37.
	LI Songsong, WANG Ningxin, ZHENG Wei, et al. Comparison of transgressive overyielding effect and plant diversity effects of annual and perennial legume-grass mixtures[J]. Chinese Journal of Plant Ecology, 2021, 45(1): 23-37.
[17]	杨婷婷. 荒漠草原生物量动态及碳储量空间分布研究——以苏尼特右旗为例[D]. 呼和浩特: 内蒙古农业大学, 2013.
	YANG Tingting. Study on biomass dynamics and spatial distribution of carbon storage in desert grassland— a case study of sunite right banner[D]. Hohhot: Inner Mongolia Agricultural University, 2013.
[18]	郭涛, 颜安, 耿洪伟. 基于无人机影像的小麦株高与LAI预测研究[J]. 麦类作物学报, 2020, 40(9): 1129-1140.
	GUO Tao, YAN An, GENG Hongwei. Prediction of wheat plant height and leaf area index based on UAV image[J]. Journal of Triticeae Crops, 2020, 40(9): 1129-1140.
[19]	刘杨, 冯海宽, 孙乾, 等. 不同分辨率无人机数码影像的马铃薯地上生物量估算研究[J]. 光谱学与光谱分析, 2021, 41(5): 1470-1476.
	LIU Yang, FENG Haikuan, SUN Qian, et al. Estimation study of above ground biomass in potato based on UAV digital images with different resolutions[J]. Spectroscopy and Spectral Analysis, 2021, 41(5): 1470-1476.
[20]	李天驰, 冯海宽, 田坤云, 等. 基于PROSAIL模型和无人机高光谱数据的冬小麦LAI反演[J]. 麦类作物学报, 2022, 42(11): 1408-1418.
	LI Tianchi, FENG Haikuan, TIAN Kunyun, et al. Retrieval of winter wheat leaf area index by PROSAIL model and hyperspectral data[J]. Journal of Triticeae Crops, 2022, 42(11): 1408-1418.
[21]	甘甜, 李雷, 李红叶, 等. 基于多源遥感数据和机器学习算法的冬小麦产量预测研究[J]. 麦类作物学报, 2022, 42(11): 1419-1428.
	GAN Tian, LI Lei, LI Hongye, et al. Winter wheat yield prediction based on multi-source remote sensing data and machine learning algorithms[J]. Journal of Triticeae Crops, 2022, 42(11): 1419-1428.
[22]	钟滨, 廖永皓, 蔡海生. 基于光谱与纹理特征的高分二号竹林信息提取——以庐山自然保护区为例[J]. 测绘与空间地理信息, 2020, 43(12): 8-13.
	ZHONG Bin, LIAO Yonghao, CAI Haisheng. Extraction bamboo forest from GF-2 satellite based on spectral and texture features: taking Lushan nature reserve as an example[J]. Geomatics & Spatial Information Technology, 2020, 43(12): 8-13.

主要参数 Main parameters	参数值 Parameter value
最大起飞重量 Maximum takeoff weight(g)	1367
续航时间 Endurance(min)	24~27
横、纵向重叠率 Horizontal and vertical overlap rates(%)	80
最大起飞海拔 Maximum takeoff altitude/m	6000
最大分辨率/(像素×像素) Maximum resolution (pixels × pixels)	4000×3000
波段类型 Band type	R、G、B

主要参数 Main parameters	参数值 Parameter value
最大起飞重量 Maximum takeoff weight(g)	1367
续航时间 Endurance(min)	24~27
横、纵向重叠率 Horizontal and vertical overlap rates(%)	80
最大起飞海拔 Maximum takeoff altitude/m	6000
最大分辨率/(像素×像素) Maximum resolution (pixels × pixels)	4000×3000
波段类型 Band type	R、G、B

序号 Serial number	10 m		30 m		50 m		70 m		90 m
序号 Serial number	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|
1	VARI	0.53	EXG	0.623	EXG	0.644	EXG	0.561	EXG	0.560
2	EXGR	0.48	VARI	0.605	EXGR	0.603	VARI	0.494	VARI	0.532
3	GRVI	0.47	EXGR	0.555	VARI	0.582	MGRVI	0.487	EXGR	0.428
4	NDI	0.35	GRVI	0.491	EXR	0.314	EXGR	0.442	GRVI	0.386
5	EXR	0.34	NDI	0.410	MGRVI	0.202	EXR	0.158	MGRVI	0.342
6	MGRVI	0.30	MGRVI	0.392	NDI	0.180	RGBVI	0.104	RGBVI	0.226
7	RGBVI	0.29	EXR	0.288	RGBVI	0.097	GRVI	0.104	NDI	0.215
8	EXG	0.25	RGBVI	0.162	GRVI	0.097	NDI	0.087	EXR	0.147

序号 Serial number	10 m		30 m		50 m		70 m		90 m
序号 Serial number	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	影像指数 Image- index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|
1	VARI	0.53	EXG	0.623	EXG	0.644	EXG	0.561	EXG	0.560
2	EXGR	0.48	VARI	0.605	EXGR	0.603	VARI	0.494	VARI	0.532
3	GRVI	0.47	EXGR	0.555	VARI	0.582	MGRVI	0.487	EXGR	0.428
4	NDI	0.35	GRVI	0.491	EXR	0.314	EXGR	0.442	GRVI	0.386
5	EXR	0.34	NDI	0.410	MGRVI	0.202	EXR	0.158	MGRVI	0.342
6	MGRVI	0.30	MGRVI	0.392	NDI	0.180	RGBVI	0.104	RGBVI	0.226
7	RGBVI	0.29	EXR	0.288	RGBVI	0.097	GRVI	0.104	NDI	0.215
8	EXG	0.25	RGBVI	0.162	GRVI	0.097	NDI	0.087	EXR	0.147

序号 Serial number	10 m		30 m		50 m		70 m		90 m
序号 Serial number	纹理指数 Texture index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	纹理指数 Texture index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	纹理指数 Texture index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	纹理指数 Texture index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|	纹理指数 Texture index	相关系数绝对值\|r\| Absolute value of correlation coefficient \|r\|
1	mea_G	0.496	mea_B	0.493	mea_B	0.466	ang_B	0.401	ent_B	0.421
2	mea_B	0.490	ang_B	0.460	con_G	0.360	ang_R	0.348	dis_G	0.352
3	cor_B	0.440	ent_G	0.455	ang_G	0.345	hom_R	0.353	ang_G	0.370
4	hom_G	0.434	hom_R	0.454	ent_B	0.312	hom_G	0.305	ent_G	0.339
5	con_B	0.425	ent_G	0.445	ent_G	0.305	dis_B	0.301	cor_R	0.316
6	dis_R	0.419	ang_R	0.361	hom_B	0.332	var_B	0.299	dis_B	0.279
7	ang_R	0.407	ent_R	0.263	ang_R	0.331	var_G	0.267	ang_R	0.254
8	ent_B	0.406	dis_B	0.257	ent_R	0.215	con_B	0.264	ent_R	0.169