Estimating of Wheat Ears Number in Field Based on RGB Images Using Unmanned Aerial Vehicle

doi:10.13304/j.nykjdb.2021.0335

Abstract

Abstract:

Ears number per unit area is the key element of wheat yield， so estimating ears number in field quickly and accurately based on aerial photographs and information extraction could provide the direct support to wheat growth monitoring and production estimates. This study acquired the unified standard and high resolution RGB aerial photographs using unmanned aerial vehicle （UAV） controlled by fragmentation monitoring and analysis with aerial photography （FragMAP）， then obtained the training model by quick and efficient target detection method （YOLOv3） and realized identifying wheat ears automatically， and finally established the estimation model of wheat ears number based on the relationships of the wheat ears number measured by proposed method （FY） and traditional methods. The results showed that the sampling efficiency and monitor area of FY were significantly higher than that of traditional method； the mean accuracy rate of identification based on YOLOv3 model was above 90%； the wheat ears numbers measured by FY and traditional method were significant linear correlation， and the estimate model was established as y=0.816x-14.863 （R²=0.790， P<0.001）. The results showed that the wheat ears estimation based on the unified standard and higher resolution images collected by UAV and deep learning was accurate， and could effectively monitor wheat growth and predicting the wheat yield.

Key words: aerial photograph, FragMAP, deep learning, YOLOv3, wheat ears detection

摘要：

单位面积穗数是小麦产量构成的重要因素，利用图像信息处理技术快速、准确地估测田间小麦穗数，可以为小麦长势监测和产量估测提供直接依据。利用无人机路径规划和控制系统（fragmentation monitoring and analysis with aerial photography， FragMAP）获取标准统一、高分辨率的田间小麦RGB航拍影像，通过高效的目标检测手段（YOLOv3）获得训练模型并自动识别麦穗，通过分析该方法（FY方法）与传统方法测定麦穗数量的关系来构建单位面积麦穗估测模型。结果表明，FY方法的样本获取效率和观测面积显著高于传统方法（P < 0.001）；YOLOv3训练模型识别麦穗的准确率随着训练样本数量和迭代次数增加而增加，500个训练样本迭代6 250次，获得模型识别麦穗的准确率超过90%；FY方法和与传统方法测定的田间小麦穗数量呈显著的线性相关关系，据此构建估测田间小麦穗数的模型为：y=0.816x-14.863（R²= 0.790，P<0.001）。上述结果表明，结合标准统一、高分辨率的无人机航拍影像和深度学习方法估测田间小麦穗数精度高、实时性强，可为小麦长势监测和产量估测提供重要的数据和技术支撑。

关键词: 航拍, FragMAP, 深度学习, YOLOv3, 麦穗检测

CLC Number:

S127

Yinyan GAO, Yi SUN, Baochun LI. Estimating of Wheat Ears Number in Field Based on RGB Images Using Unmanned Aerial Vehicle[J]. Journal of Agricultural Science and Technology, 2022, 24(3): 103-110.

高姻燕, 孙义, 李葆春. 基于无人机RGB影像估测田间小麦穗数[J]. 中国农业科技导报, 2022, 24(3): 103-110.

Figures/Tables 6

Fig.1 Study area and filed sampling plots

Fig.2 Collection of wheat ears information by UAV

Fig.3 Field sampling efficiency of FY and traditional methodsNote：*** indicates significant difference between methods at P<0.001 level.

Fig.4 Examples of the recognition by the 3 trained models

Fig.5 Relationship between recognition accuracy of training models and sample number and iterationsNote： Different letters indicate significant differences at P<0.05 level.

Fig.6 Prediction model based on the data collected using traditional and FY methodNote： Dashed line indicates line 1∶1.

References 37

1	赵广才,常旭虹,王德梅,等.小麦生产概况及其发展[J].作物杂志,2018,185(4):1-7.
	ZHAO G C, CHANG X H, WANG D M, et al.. General situation and development of wheat production [J]. Crops, 2018, 185(4):1-7.
2	杜颖,蔡义承,谭昌伟,等.基于超像素分割的田间小麦穗数统计方法[J].中国农业科学,2019,52(1):21-33.
	DU Y, CAI Y C, TAN C W, et al.. Field wheat ears counting based on superpixel segmentation method [J]. Sci. Agric. Sin., 2019, 52(1):21-33.
3	王永春,李静,王秀东.新中国成立以来我国粮食生产变动规律研究及趋势展望[J].中国农业科技导报,2021,23(1):1-11.
	WANG Y C, LI J, WANG X D. Change law and trend of gain production in china since the founding of the People’s Republic of China [J]. J. Agric. Sci. Technol., 2021, 23(1):1-11.
4	丁国辉,许昊,温明星,等.基于经济型低空无人机对小麦重要产量表型性状的多生育时期获取和自动化分析[J].农业大数据学报,2019,1(2):19-31.
	DING G H, XU H, WEN M X, et al.. Developing cost-effective and low-altitude UAV aerial phenotyping and automated phenotypic analysis to measure key yield-related traits for bread wheat [J]. J. Agric. Big Data, 2019, 1(2):19-31.
5	FENGMEI G, WEIE W, JINDONG L, et al.. Genome-wide linkage mapping of QTL for yield components, plant height and yield-related physiological traits in the Chinese wheat cross zhou 8425B/Chinese spring [J/OL]. Front. Plant Sci., 2020, 6:1099 [2021-12-15]. .
6	EDAE E A, BYRNE P, HALEY S, et al.. Genome-wide association mapping of yield and yield components of spring wheat under contrasting moisture regimes [J]. Theor. Appl. Genet., 2014, 127(4):791-807.
7	ZHAO R, ZHAO A. Chinese main crops yield estimate by remote sensing [C]// Proceedings of the Euro-Asia Space Week on Cooperation in Spece. 1999, 430:275.
8	刘欣谊,仲晓春,陈晨,等.利用无人机图像颜色与纹理特征数据在小麦生育前期对产量进行预测[J].麦类作物学报,2020,40(8):1002-1007.
	LIU X Y, ZHONG X C, CHEN C, et al.. Prediction of wheat yield using color and texture feature data of UAV image at early growth stage [J]. J. Triticeae Crops, 2020, 40(8):1002-1007.
9	周清波.国内外农情遥感现状与发展趋势[J].中国农业资源与区划,2004,25(5):9-14.
	ZHOU Q B. Status and tendency for development in remote sensing of agriculture situation [J]. Chin. J. Agric. Resour. Reg. Plan., 2004, 25(5):9-14.
10	吕书强,晏磊,张兵,等.无人机遥感系统的集成与飞行试验研究[J].测绘科学,2007,32(1):84-86.
	LYU S Q, YAN L, ZHANG B, et al.. The integration and flight experiment of UAV remote sensing systems [J]. Sci. Surv. Map., 2007, 32(1):84-86.
11	BARETH G, AASEN H, BENGIG J, et al.. Low-weight and UAV-based hyperspectral full-frame cameras for monitoring crops: spectral comparison with portable spectroradiometer measurements [J]. Photogramm Fernerkun, 2015, (1):69-79.
12	GENTLE M, FINCH N, SPEED J, et al.. A comparison of unmanned aerial vehicles (drones) and manned helicopters for monitoring macropod populations [J]. Wildlife Res., 2018, 45(7):586-594.
13	CHRISTIE K S, GILBERT S L, BROWN C L, et al.. Unmanned aircraft systems in wildlife research: current and future applications of a transformative technology [J]. Front. Ecol. Environ., 2016, 14(5):241-251.
14	YI S. FragMAP: a tool for long-term and cooperative monitoring and analysis of small-scale habitat fragmentation using an unmanned aerial vehicle [J]. Int. J. Remote Sens., 2017, 38(8-10):2686-2697.
15	刘刚,许宏健,马海涛,等.无人机航测系统在应急服务保障中的应用与前景[J].测绘与空间地理信息,2011,34(4):177-179.
	LIU G, XU H J, MA H T, et al.. Unmanned aerial aystem applications and prospects in the protection of emergency services [J]. Geom. Spatial Inform. Tech., 2011, 34(4):177-179.
16	杨贵军,李长春,于海洋,等.农用无人机多传感器遥感辅助小麦育种信息获取[J].农业工程学报,2015,31(21):184-90.
	YANG G J, LI C C, YU H Y, et al.. UAV based multi-load remote sensing technologies for wheat breeding information acquirement [J]. Trans. Chin. Soc. Agric. Eng., 2015, 31(21):184-90.
17	李艳花.提高冬小麦分蘖成穗率的几项关键措施[J].现代农村科技,2018,11:18.
18	刘哲,袁冬根,王恩.基于改进Bayes抠图算法的麦穗小穗自动计数方法[J].中国农业科技导报,2020,22(8):75-82.
	LIU Z, YUAN D G, WANG E. Automatic counting method of wheat grain based on improved Bayes matting algorithm [J]. J. Agric. Sci. Technol., 2020, 22(8):75-82.
19	田纪春,邓志英,胡瑞波,等.不同类型超级小麦产量构成因素及籽粒产量的通径分析[J].作物学报,2006,32(11):1699-1705.
	TIAN J C, DENG Z Y, HU R B, et al.. Yield components of super wheat cultivars with different types and the path coefficient analysis on grain yield [J]. Acta Agron. Sin, 2006, 32(11):1699-1705.
20	方正,邵锡珍,李云海.从小麦的超高产实践谈良种选育的问题[J].作物杂志,1999,(1):20-21.
	FANG Z, SHAO X Z, LI Y H. Discuss breeding based on the super-high-yield practice of wheat [J]. Crops, 1999, 1: 20-21.
21	高宇,高军萍,李寒,等.植物表型监测技术研究进展及发展对策[J].江苏农业科学,2017,45(11):5-11.
	GAO Y, GAO J P, LI H, et al.. Research progress and development countermeasures [J]. Jiangsu Agric. Sci., 2017, 45(11): 5-11.
22	陈含,吕行军,田凤珍,等.基于Sobel算子边缘检测的麦穗图像分割[J].农机化研究,2013,35(3):33-36.
	CHEN H, LYU X J, TIAN F Z, et al.. Wheat panicle image segmentation based on Sobel operator-edge detection [J]. J. Agric. Mech. Res., 2013, 35(3):33-36.
23	路文超,罗斌,潘大宇,等.基于图像处理的小麦穗长和小穗数同步测量[J].中国农机化学报,2016, 37(6):210-215.
	LU W C, LUO B, PAN D Y, et al.. Synchronous measurement of wheat ear length and spikelets number based on image processing [J]. J. Chin. Agric. Mech., 2016, 37(6):210-215.
24	LI Q, CAI J, BERGER B, et al.. Detecting spikes of wheat plants using neural networks with Laws texture energy [J/OL]. Plant Methods, 2017, 13(1):83 [2021-12-15]. .
25	刘涛,孙成明,王力坚,等.基于图像处理技术的大田麦穗计数[J].农业机械学报,2014,45(2):232-290.
	LIU T, SUN C M, WANG L J, et al.. In-field wheatheat counting based on image processing technology [J]. Trans. Chin. Soc. Agric. Machinery, 2014, 45(2):232-290.
26	魏宏彬,张端金,杜广明,等.基于改进型YOLO v3的蔬菜识别算法[J].郑州大学学报,2020,41(2):7-12, 31.
	WEI H B, ZHANG D J, DU G M, et al.. Vegetable recognition algorithm based on improved YOLOv3 [J]. J. Zhengzhou Univ., 2020, 41(2):7-12, 31.
27	任嘉锋,熊卫华,吴之昊,等.基于改进YOLOv3的火灾检测与识别[J].计算机系统应用,2019,28(12):171-176.
	REN J F, XIONG W H, WU Z H, et al.. Fire detection and identification based on improved YOLOv3 [J]. Comput. Syst. Appl., 2019, 28(12):171-176.
28	崔博超,郑江华,刘忠军,等.无人机遥感影像的YOLOv3鼠洞识别技术[J].林业科学,2020,56(10):199-208.
	CUI B C, ZHENG J H, LIU Z J, et al.. YOLOv3 mouse hole recognition based on remote sensing images from technology for unmanned aerial vehicle [J]. Sci. Silv. Sin., 2020, 56(10):199-208.
29	范祥妍.临夏县冬小麦高产栽培技术[J].新农村,2018(26):109.
30	何苏琴,金秀琳,鲁振超.甘肃省临夏州小麦脚腐病病原鉴定[J].植物保护,2006,32(3):35-38.
	HE S Q, JIN X L, LU Z C. Isolation and identification of the pathogens causing wheat foot rot in Linxia region of Gansu Province [J]. Plant Prot., 2006, 32(3):35-38.
31	王长耀,林文鹏.基于MODIS EVI的冬小麦产量遥感预测研究[J].农业工程学报,2005,21(10):90-94.
	WANG C Y, LIN W P. Winter wheat yield estimation based on MODIS EVI [J]. Trans. Chin. Soc. Agric. Eng., 2005, 21(10):90-94.
32	BENDIG J, BOLTEN A, BARETH G. UAV-based imaging for multi-temporal, very high resolution crop surface models to monitor crop growth variability [J]. Photogramm. Fernerkun., 2013 (6):551-562.
33	JANNOURA R, BRINKMANN K, UTEAU D, et al.. Monitoring of crop biomass using true colour aerial photographs taken from a remote controlled hexacopter [J]. Biosyst. Eng., 2015, 129:341-351.
34	周恺,周彤,丁峰,等.基于无人机图像的小麦主要生育时期LAI估算[J].中国农业科技导报,2021,23(1):89-97.
	ZHOU K, ZHOU T, DING F, et al.. Wheat LAI estimation in main growth period based on UAV images [J]. J. Agric. Sci. Technol., 2021, 23(1):89-97.
35	ZHANG J G, LIU D W, MENG B P, et al.. Using UAVs to assess the relationship between alpine meadow bare patches and disturbance by pikas in the source region of Yellow River on the Qinghai-Tibetan Plateau [J/OL]. Glob. Ecol. Conserv., 2021, 26(6):e01517[2021-12-16]. .
36	高露,马元婧.基于Faster R-CNN的设备故障检测与识别[J].计算机系统应用,2019,28(4):170-175.
	GAO L, MA Y J. Equipment fault detection and recognition based on faster R-CNN [J]. Comput. Syst. Appl., 2019, 28(4):170-175.
37	WIEDERHOLD B K. Citizen scientists generate benefits for researchers, educators, society, and themselves [J]. Cyberpsychol. Behav. Soc. Netv., 2011, 14(12):703-704.

[1]	ZHOU Huiru, WU Boming. Advances in Research on Deep Learning for Crop Disease Image Recognition [J]. Journal of Agricultural Science and Technology, 2021, 23(5): 61-68.
[2]	LYU Chunyang, LIU Shengping, GUO Xiuming, XIAO Shunfu, LIU Dazhong, YANG Feifei, LI Luhua. Detection of Honeybee Based on SSD Model [J]. Journal of Agricultural Science and Technology, 2021, 23(5): 98-107.
[3]	GUO Yang, XU Beibei, CHEN Guipeng, DING Jian, YAN Zhiyan, LIANG Hua, WU Changhua. Rice Insect Pest Recognition Method Based on Convolutional Neural Network [J]. Journal of Agricultural Science and Technology, 2021, 23(11): 99-109.
[4]	XU Beibei1, WANG Wensheng1,2*, GUO Leifeng1, CHEN Guipeng3. A Review and Future Prospects on Cattle Recognition Based on Non-contact Identification [J]. Journal of Agricultural Science and Technology, 2020, 22(7): 79-89.
[5]	DONG Wei1, QIAN Rong1, ZHANG Jie 2*, ZHANG Liping1, CHEN Hongbo2, ZHANG Meng1, ZHU Jingbo1, BU Yingqiao3. Vegetable Lepidopteran Pest Auto Recognition and Detection Counting Based on Deep Learning [J]. Journal of Agricultural Science and Technology, 2019, 21(12): 76-84.