Study on extraction of apple tree crown width at different flight altitudes using multi-spectral UAV

doi:10.6048/j.issn.1001-4330.2024.06.021

Abstract

Abstract:

【Objective】 The objective of this study is to segment and extract the crown information by using the multi-spectral image of the UAV, so as to realize the monitoring of the apple tree crown information, and analyze the influence of the UAV flight height on the extraction accuracy of the crown. 【Methods】 The DJI Phantom 4 multi-spectral UAV was used to capture multi-spectral images of apple trees at flight altitudes of 30, 60, and 90 m. The acquired images were processed using DJI Terra software to generate Digital Orthophoto Maps (DOM) and Digital Surface Models (DSM). The multi-scale segmentation method and threshold-based classification were utilized to segment the apple tree crowns and extract the east-west and north-south crown width values of the target trees. The accuracy of the segmentation and extraction results was validated by comparing them with visual interpretation of segmented crowns and field measurements of crown widths for 100 apple trees. 【Results】 At a flight altitude of 30 m, the accuracy of individual crown segmentation was 84.00%. The average extraction accuracy of the east-west and north-south crown widths was 91.15%. The average R-squared value (R²) was 0.7859, and the average root mean square error (RMSE) was 0.216 6 m. At a flight altitude of 60m, the accuracy of individual crown segmentation was 81.00%. The average extraction accuracy of the crown widths was 90.31%. The average R² was 0.7376, and the average RMSE was 0.241 6 m. At a flight altitude of 90m, the accuracy of individual crown segmentation was 73.00%. The average extraction accuracy of the crown widths was 88.88%. The average R² was 0.710 2, and the average RMSE was 0.267 6 m. 【Conclusion】 The study concludes that the use of multi-spectral UAV remote sensing technology enables the monitoring of apple tree crown information possible. The results also indicate that flight altitude has a certain impact on the extraction accuracy. As the UAV flight altitude increases, the accuracy of individual crown segmentation and extraction of crown widths gradually decrease. UAV remote sensing technology offers the advantages of low cost, rapid data acquisition, and high accuracy, which can significantly improve the efficiency of orchard management for fruit growers.

Key words: multi-spectral; UAV; flight altitude; apple tree; crown

摘要：

【目的】利用无人机多光谱影像分割并提取苹果树树冠冠幅信息,实现无人机遥感技术监测苹果树树冠信息,并分析无人机飞行高度对冠幅提取精度的影响。【方法】利用大疆精灵4多光谱无人机分别获取30、60和90m飞行高度的苹果树多光谱影像,经大疆智图(DJI Terra)软件处理生成DOM和DSM影像数据,利用多尺度分割法和阈值分类对苹果树树冠进行分割并提取目标树的东西、南北冠幅值,以目视解译分割树冠和实地测量的100株果树冠幅值为参考进行精度验证。【结果】30 m飞行高度的单木树冠分割准确率为84.00%,东西、南北冠幅平均提取精度为91.15%,平均R²为0.785 9,平均RMSE为0.216 6 m;60 m飞行高度的单木树冠分割准确率为81.00%,东西、南北冠幅平均提取精度为90.31%,平均R²为0.737 6,平均RMSE为0.241 6 m;90 m飞行高度的单木树冠分割准确率为73.00%,东西、南北冠幅平均提取精度为88.88%,平均R²为0.710 2, 平均RMSE为0.267 6 m。【结论】利用多光谱无人机遥感技术可以实现对苹果树树冠信息的监测;飞行高度对提取结果有一定的影响,随着无人机飞行高度的增加,单木分割准确率和冠幅提取精度逐渐降低。无人机遥感技术成本低、快速、准确,可以提高果园的管理效率。

关键词: 多光谱, 无人机, 飞行高度, 苹果树, 树冠

CLC Number:

S661.1

ZHANG Zhenfei, GUO Jing, YAN An, YUAN Yilin, XIAO Shuting, HOU Zhengqing, SUN Zhe. Study on extraction of apple tree crown width at different flight altitudes using multi-spectral UAV[J]. Xinjiang Agricultural Sciences, 2024, 61(6): 1468-1476.

张振飞, 郭靖, 颜安, 袁以琳, 肖淑婷, 侯正清, 孙哲. 基于多光谱无人机不同飞行高度下苹果树冠幅信息的提取[J]. 新疆农业科学, 2024, 61(6): 1468-1476.

Figures/Tables 8

Fig.1 UAV image stitching process

Fig.2 Tree crown area extraction results at 30 m flight altitude

Fig.3 Five segmentation situations Note:Match A;B is close to matching;C is lost;D over segmentation; Under segmentation of E;Blue as reference tree crown; Red indicates automatic segmentation of tree crown

Fig.4 Optimal scale line char

Fig.5 Automatic tree crown segmentation results at different drone flight heights Note:a:Automatic segmentation a flight altitude of 30 m;b:Automatic segmentation at a flight altitude of 60 m;c:Automatic segmentation at a flight altitude of 90 m

Tab.1 Tree crown segmentation statistics

飞行高度 Flying altitude (m)	匹配率 Matching rate (%)	接近匹配率 Close matching rate (%)	丢失率 Loss rate (%)	过分割率 Over-segmentation rate(%)	欠分割率 Under-segmentation rate(%)
30 60 90	47.00 39.00 29.00	37.00 42.00 44.00	4.00 4.00 4.00	5.00 7.00 9.00	7.00 8.00 14.00

Tab.2 Regression statistics of extracted crown width and measured crown width

飞行高度 Flying altitude (m)	精度范围 Accuracy range (%)	平均提取精度 Average extraction accuracy(%)	R²	RMSE(m)
30-EW 30-NS 60-EW 60-NS 90-EW 90-NS	72.99~99.90 79.40~98.99 70.80~98.91 78.74~98.99 71.53~99.47 41.36~98.96	90.43 91.86 89.54 91.09 89.16 88.60	0.770 2 0.801 5 0.724 3 0.750 9 0.705 6 0.714 8	0.204 7 0.228 5 0.226 4 0.256 7 0.228 9 0.306 3

Fig.6 Different flight altitudes scatter plot of average extracted crown width and measured values Note:Blue,red,green respectively 30,60,90 m scatter plot of fight altitude

References 28

[1]	赵祎, 马力群, 罗云波, 等. 贮藏温度对嘎啦苹果贮藏期及货架期品质的影响[J]. 保鲜与加工, 2018, 18(3): 8-14.
	ZHAO Yi, MA Liqun, LUO Yunbo, et al. Influence of storage temperature on quality of gala apple during storage and shelf life[J]. Storage and Process, 2018, 18(3): 8-14.
[2]	郭靖, 韩政伟, 张振飞, 等. 苹果新品种王林1号的选育[J]. 果树学报, 2023, 40(2): 386-389.
	GUO Jing, HAN Zhengwei, ZHANG Zhenfei, et al. Breeding report of a new apple cultivar Wanglin 1[J]. Journal of Fruit Science, 2023, 40(2): 386-389.
[3]	覃先林, 李增元, 易浩若. 高空间分辨率卫星遥感影像树冠信息提取方法研究[J]. 遥感技术与应用, 2005, 20(2): 228-232.
	QIN Xianlin, LI Zengyuan, YI Haoruo. Extraction method of tree crown using high-resolu tion satellite image[J]. Remote Sensing Technology and Application, 2005, 20(2): 228-232.
[4]	史云, 沈磊, 陈山, 等. 基于低空遥感的果树冠层信息提取方法研究[J]. 中国农业信息, 2022, 34(1): 1-10.
	SHI Yun, SHEN Lei, CHEN Shan, et al. Research on the extraction method of fruit tree canopy information based on low-altitude remote sensing[J]. China Agricultural Informatics, 2022, 34(1): 1-10.
[5]	万祖毅, 田永中, 刘旭东, 等. 基于轻小型无人机影像的柑橘树株数提取[J]. 湖北农机化, 2020,(7): 140-144.
	WAN Zuyi, TIAN Yongzhong, LIU Xudong, et al. Extracting the Number of Citrus Trees Based on Light and Small UAV Images[J]. Hubei Agricultural Mechanization, 2020,(7): 140-144.
[6]	祁媛, 徐伟诚, 王林琳, 等. 基于无人机遥感影像的沙糖橘果树提取方法研究[J]. 华南农业大学学报, 2020, 41(6): 126-133.
	QI Yuan, XU Weicheng, WANG Linlin, et al. Study on the extraction method of sugar tangerine fruit trees based on UAV remote sensing images[J]. Journal of South China Agricultural University, 2020, 41(6): 126-133.
[7]	Yang H B, Zhao J, Lan Y B, et al. Fraction vegetation cover extraction of winter wheat based on RGB image obtained by UAV[J]. International Journal of Precision Agricultural Aviation, 2018, 1(1): 54-61.
[8]	牛亚晓, 张立元, 韩文霆, 等. 基于无人机遥感与植被指数的冬小麦覆盖度提取方法[J]. 农业机械学报, 2018, 49(4): 212-221.
	NIU Yaxiao, ZHANG Liyuan, HAN Wenting, et al. Fractional vegetation cover extraction method of winter wheat based on UAV remote sensing and vegetation index[J]. Transactions of the Chinese Society for Agricultural Machinery, 2018, 49(4): 212-221.
[9]	颜安, 郭涛, 陈全家, 等. 基于无人机影像的棉花株高预测[J]. 新疆农业科学, 2020, 57(8): 1493-1502. DOI
	YAN An, GUO Tao, CHEN Quanjia, et al. Prediction of cotton plant height based on UAV image[J]. Xinjiang Agricultural Sciences, 2020, 57(8): 1493-1502. DOI
[10]	周辰琴, 余拥军, 方陆明, 等. 飞行高度与郁闭度对水杉冠幅提取影响的研究[J]. 林业资源管理, 2022,(1): 150-156.
	ZHOU Chenqin, YU Yongjun, FANG Luming, et al. Effects of flight height and canopy density on canopy extraction of Metasequoia glyptostroboides[J]. Forest Resources Management, 2022,(1): 150-156.
[11]	Wang L, Gong P, Biging G S. Individual tree-crown delineation and treetop detection in high-spatial-resolution aerial imagery[J]. Photogrammetric Engineering & Remote Sensing, 2004, 70(3): 351-357.
[12]	郭昱杉, 刘庆生, 刘高焕, 等. 基于标记控制分水岭分割方法的高分辨率遥感影像单木树冠提取[J]. 地球信息科学学报, 2016, 18(9): 1259-1266. DOI
	GUO Yushan, LIU Qingsheng, LIU Gaohuan, et al. Individual tree crown extraction of high resolution image based on marker-controlled watershed segmentation method[J]. Journal of Geo-Information Science, 2016, 18(9): 1259-1266.
[13]	徐永胜, 刘浩然, 范伟伟, 等. 基于分水岭算法的无人机不同飞行高度下林木冠幅提取研究[J]. 西北林学院学报, 2021, 36(3): 197-202, 259.
	XU Yongsheng, LIU Haoran, FAN Weiwei, et al. Extraction of forest stand crown width at different flying heights of UAV based on watershed algorithm[J]. Journal of Northwest Forestry University, 2021, 36(3): 197-202, 259.
[14]	孙钊, 潘磊, 孙玉军. 基于无人机影像的高郁闭度杉木纯林树冠参数提取[J]. 北京林业大学学报, 2020, 42(10): 20-26.
	SUN Zhao, PAN Lei, SUN Yujun. Extraction of tree crown parameters from high-density pure Chinese fir plantations based on UAV images[J]. Journal of Beijing Forestry University, 2020, 42(10): 20-26.
[15]	万祖毅. 基于无人机遥感的柑橘果树信息提取及应用研究[D]. 重庆: 西南大学, 2020.
	WAN Zuyi. Extraction and Application of Citrus Fruit Tree Information Based on UAV Remote Sensing[D]. Chongqing: Southwest University, 2020.
[16]	王雅佩, 王振锡, 李擎, 等. 基于无人机影像的天山云杉林伐后更新地林分密度估测研究[J]. 中南林业科技大学学报, 2020, 40(5): 57-66.
	WANG Yapei, WANG Zhenxi, LI Qing, et al. Research on estimation of stand density of regenerate Picea schrenkiana var. tianshanica based on UAV images[J]. Journal of Central South University of Forestry & Technology, 2020, 40(5): 57-66.
[17]	郭旭展, 陈巧, 张晓芳, 等. 基于无人机高分辨率影像的油松新造林健康树冠提取[J]. 林业科学, 2022, 58(10): 111-120.
	GUO Xuzhan, CHEN Qiao, ZHANG Xiaofang, et al. Extraction of healthy canopy of new afforestation for Pinus tabulaeformis based on UAV high-resolution image[J]. Scientia Silvae Sinicae, 2022, 58(10): 111-120.
[18]	刘方舟, 刘浩, 云挺. 基于分水岭优化思想的单木信息分割算法[J]. 林业工程学报, 2020, 5(5): 109-116.
	LIU Fangzhou, LIU Hao, YUN Ting. Individual tree crown separation using the improved watershed method[J]. Journal of Forestry Engineering, 2020, 5(5): 109-116.
[19]	Erikson M. Segmentation of individual tree crowns in colour aerial photographs using region growing supported by fuzzy rules[J]. Canadian Journal of Forest Research, 2003, 33(8): 1557-1563.
[20]	Jing L H, Hu B X, Li J L, et al. Automated delineation of individual tree crowns from lidar data by multi-scale analysis and segmentation[J]. Photogrammetric Engineering & Remote Sensing, 2012, 78(12): 1275-1284.
[21]	王枚梅, 林家元, 林沂, 等. 基于无人机可见光影像的亚高山针叶林树冠参数信息自动提取[J]. 林业资源管理, 2017,(4): 82-88.
	WANG Meimei, LIN Jiayuan, LIN Yi, et al. Subalpine coniferous forest crown information automatic extraction based on optical UAV remote sensing imagery[J]. Forest Resources Management, 2017,(4): 82-88.
[22]	孙钊. 基于无人机航拍图像的杉木树冠参数提取与模型研建[D]. 北京: 北京林业大学, 2020.
	SUN Zhao. Extraction of Tree Crown Parameters and Model Developmentbased on UAV Image[D]. Beijing: Beijing Forestry University, 2020.
[23]	李文静, 王瑞瑞, 石伟, 等. 基于无人机多光谱影像的单木树冠提取方法[J]. 福建农林大学学报(自然科学版), 2020, 49(5): 639-645.
	LI Wenjing, WANG Ruirui, SHI Wei, et al. Crown extraction method of individual tree based on multi-spectral image of UAV[J]. Journal of Fujian Agriculture and Forestry University (Natural Science Edition), 2020, 49(5): 639-645.
[24]	束美艳, 李世林, 魏家玺, 等. 基于无人机平台的柑橘树冠信息提取[J]. 农业工程学报, 2021, 37(1): 68-76.
	SHU Meiyan, LI Shilin, WEI Jiaxi, et al. Extraction of citrus crown parameters using UAV platform[J]. Transactions of the Chinese Society of Agricultural Engineering, 2021, 37(1): 68-76.
[25]	张玉薇, 陈棋, 田湘云, 等. 基于UAV可见光遥感的单木冠幅提取研究[J]. 西部林业科学, 2022, 51(3): 49-59.
	ZHANG Yuwei, CHEN Qi, TIAN Xiangyun, et al. Individual tree crown extraction based on UAV visible light remote sensing technology[J]. Journal of West China Forestry Science, 2022, 51(3): 49-59.
[26]	于旭宅, 王瑞瑞, 陈伟杰. 改进分水岭算法在无人机遥感影像树冠分割中的应用[J]. 福建农林大学学报(自然科学版), 2018, 47(4): 428-434.
	YU Xuzhai, WANG Ruirui, CHEN Weijie. Forest canopy segmentation of UAV remote sensing images using improved watershed algorithm[J]. Journal of Fujian Agriculture and Forestry University (Natural Science Edition), 2018, 47(4): 428-434.
[27]	金忠明, 曹姗姗, 王蕾, 等. 基于U-Net和分水岭算法的无人机单木树冠提取方法[J]. 西北林学院学报, 2020, 35(6): 194-204.
	JIN Zhongming, CAO Shanshan, WANG Lei, et al. A method for individual tree-crown extraction from UAV remote sensing image based on U-net and watershed algorithm[J]. Journal of Northwest Forestry University, 2020, 35(6): 194-204.
[28]	郑鑫, 王瑞瑞, 靳茗茗. 基于形态学阈值标记分水岭算法的高分辨率影像单木树冠提取[J]. 中南林业调查规划, 2017, 36(4): 30-35, 57.
	ZHENG Xin, WANG Ruirui, JIN Mingming. Extraction of high-resolution images of single tree crown based on watershed algorithm with morphological threshold mark[J]. Central South Forest Inventory and Planning, 2017, 36(4): 30-35, 57.