Crops ›› 2023, Vol. 39 ›› Issue (2): 245-252.doi: 10.16035/j.issn.1001-7283.2023.02.035

Previous Articles    

Unmanned Aerial Vehicle Hyperspectral Estimation of Nitrogen Content in Cotton Leaves Based on RF and SPA

Yi Xiang1(), Lü Xin1(), Zhang Lifu1,3, Tian Min2, Zhang Ze1, Fan Xianglong1   

  1. 1Agriculture College of Shihezi University/Key Laboratory of Oasis Eco-Agriculture, Xinjiang Production and Construction Corps, Shihezi 832003, Xinjiang, China
    2College of Mechanical and Electrical Engineering, Shihezi University, Shihezi 832003, Xinjiang, China
    3Aerospace Information Research Institute, Chinese Academy of Sciences/State Key Laboratory of Remote Sensing Science, Beijing 100094, China
  • Received:2021-09-22 Revised:2021-10-23 Online:2023-04-15 Published:2023-04-11

RichHTML

9

PDF (PC)

212

Abstract:

In order to analysis the relationship between cotton leaf nitrogen content (LNC) and canopy spectral reflection characteristics, and achieve rapid, accurate and nondestructive monitoring of nitrogen level in the process of crop growth, based on the cotton plot experiment in Xinjiang Shihezi University Teaching and Experimental Ground in 2019, multiplicative scatter correction, smoothing algorithm, standard normal variate and first derivative were used to pretreat the original spectra of the cotton canopy. Random frog (RF) algorithm and successive projections algorithm (SPA) were used to select characteristic wavelengths and combine with partial least squares regression to establish a spectral prediction model of nitrogen content in cotton leaves. RF and SPA algorithms selected five sensitive characteristic bands of LNC from the cotton canopy spectra of 398-1000nm, and the number of bands decreased by 93.0%-96.3%, which effectively reduced the spectral redundancy information. The determination coefficient and root mean square error of partial least squares regression of LNC constructed based on SPA algorithm screening sensitive band were 0.52 and 2.55, respectively, and the determination coefficient and root mean square error of model verification were 0.70 and 2.37, respectively. The model had good accuracy and stability. It can be used as a method to estimate nitrogen content in cotton leaves by unmanned aerial vehicle.

Key words: Cotton, Unmanned aerial vehicle (UAV), Hyperspectral, Random frog, Successive projections algorithm, Partial least squares regression

Fig.1

Canopy hyperspectral reflectance at different nitrogen levels"

Fig.2

Correlation between spectral reflectance and LNC in cotton canopy"

Fig.3

Characteristic wavelength screening results of RF and SPA algorithms"

Table 1

The effective wavelengths selection for hyperspectral sample"

预处理方法
Pretreatment		 提取算法
Extraction
algorithm		 特征波长数量
Number of characteristic
wavelengths		 特征波长
Characteristic wavelength (nm)
原始光谱Original spectrum RF 10 400、445、516、520、556、669、700、705、722、898
SPA 10 403、414、676、762、953、967、971、980、996、1000
MSC RF 10 518、705、707、722、856、862、898、907、918、598
SPA
 18
 400、518、538、589、629、662、762、791、862、916、924、947、956、962、964、971、976、987
SG RF 10 400、485、516、669、691、700、705、722、845、598
SPA
 19
 398、403、407、412、414、427、516、556、669、720、938、964、967、971、976、980、993、996、1000
SNV RF 10 503、520、536、556、598、705、747、827、860、913
SPA 10 498、762、940、942、953、962、967、982、993、1000
FD RF 10 467、491、511、556、558、583、600、609、622、463
SPA
 17
 733、736、760、933、936、942、949、953、962、969、978、980、982、989、991、993、998

Table 2

Analysis of LNC estimation effect of PLSR in cotton leaves"

处理
Treatment		 训练集Train set 验证集Test set
R2 RMSE R2 RMSE
原始-RF Origin-RF 0.57 2.73 0.47 2.78
原始-SPA Origin-SPA 0.42 2.87 0.56 2.80
MSC-RF 0.53 2.59 0.57 2.73
MSC-SPA 0.46 2.81 0.60 2.57
SG-RF 0.48 2.72 0.61 2.55
SG-SPA 0.50 2.67 0.55 2.76
SNV-RF 0.54 2.56 0.58 2.67
SNV-SPA 0.41 2.90 0.52 2.85
FD-RF 0.53 2.78 0.68 2.67
FD-SPA 0.52 2.55 0.70 2.37

Fig.4

Comparison of measured values and estimated values of RF-PLSR model under different pretreatments"

Fig.5

Comparison of measured values and estimated values of SPA-PLSR model under different pretreatments"

Fig.6

Hyperspectral image and LNC inversion distribution map of cotton field"

[1] 李鹏程, 郑苍松, 孙淼, 等. 壤全氮含量和施氮水平对棉花产量及氮肥利用效率的影响. 核农学报, 2017, 31(8):1612-1617.
doi: 10.11869/j.issn.100-8551.2017.08.1612
[2] 冯帅, 许童羽, 于丰华, 等. 基于无人机高光谱遥感的东北粳稻冠层叶片氮素含量反演方法研究. 光谱学与光谱分析, 2019, 39(10):3281-3287.
[3] 赵春江. 农业遥感研究与应用进展. 农业机械学报, 2014, 45(12):277-293.
[4] 叶春, 刘莹, 李艳大, 等. 基于RGB颜色空间的早稻氮素营养监测研究. 中国农业大学学报, 2020, 25(8):25-34.
[5] Tao H, Feng H, Xu L, et al. Estimation of crop growth parameters using UAV-Based hyperspectral remote sensing data. Sensors, 2020, 20(5):1296.
doi: 10.3390/s20051296
[6] Yue J, Yang G, Tian Q, et al. Estimate of winter-wheat above-ground biomass based on UAV ultrahigh- ground-resolution image textures and vegetation indices. ISPRS Journal of Photogrammetry and Remote Sensing, 2019, 150:226-244.
doi: 10.1016/j.isprsjprs.2019.02.022
[7] 田婷, 张青, 张海东. 无人机遥感在作物监测中的应用研究进展. 作物杂志, 2020(5):1-8.
[8] 张凝, 杨贵军, 赵春江, 等. 作物病虫害高光谱遥感进展与展望. 遥感学报, 2021, 25(1):403-422.
[9] 第五鹏瑶, 卞希慧, 王姿方, 等. 光谱预处理方法选择研究. 光谱学与光谱分析, 2019, 39(9):2800-2806.
[10] 姚霞, 朱艳, 田永超, 等. 小麦叶层氮含量估测的最佳高光谱参数研究. 中国农业科学, 2009, 42(8):2716-2725.
[11] 李粉玲, 常庆瑞. 基于连续统去除法的冬小麦叶片全氮含量估算. 农业机械学报, 2017, 48(7):174-179.
[12] 梁栋, 杨勤英, 黄文江, 等. 基于小波变换与支持向量机回归的冬小麦叶面积指数估算. 红外与激光工程, 2015, 44(1):335-340.
[13] 孙晶京, 杨武德, 冯美臣, 等. 基于随机蛙跳和支持向量机的冬小麦叶面积指数估算. 山西农业大学学报(自然科学版), 2020, 40(5):120-128.
[14] Yun Y H, Li H D,E. Wood L R, et al. An efficient method of wavelength interval selection based on random frog for multivariate spectral calibration. Spectrochimica Acta Part A:Molecular and Biomolecular Spectroscopy, 2013, 111:31-36.
doi: 10.1016/j.saa.2013.03.083
[15] 朱淑鑫, 杨宸, 顾兴健, 等. K均值算法结合连续投影算法应用于土壤速效钾含量的高光谱分析. 江苏农业学报, 2020, 36(2):358-365.
[16] 吴迪, 金春华, 何勇. 基于连续投影算法的光谱主成分组合优化方法研究. 光谱学与光谱分析, 2009, 29(10):2734-2737.
[17] Geladi P, Kowalski B R. Partial least-squares regression: a tutorial. Analytica Chimica Acta, 1986, 185:1-17.
doi: 10.1016/0003-2670(86)80028-9
[18] 王惠文, 朱韵华. PLS回归在消除多重共线性中的作用. 数理统计与管理, 1996(6):48-52.
[19] 姚霞, 吴华兵, 朱艳, 等. 棉花功能叶片色素含量与高光谱参数的相关性研究. 棉花学报, 2007, 19(4):267-272.
[20] 王克如, 潘文超, 李少昆, 等. 不同施氮量棉花冠层高光谱特征研究. 光谱学与光谱分析, 2011, 31(7):1868-1872.
[21] 马文君, 常庆瑞, 田明璐, 等. 棉花全生育期叶片SPAD值的遥感估算模型. 干旱地区农业研究, 2017, 35(5):42-48.
[22] 郑一力, 赵燕东, 董玮, 等. 竹叶片氮含量高光谱估测方法对比研究. 农业机械学报, 2018, 49(增1):393-400.
[23] 王丽凤, 张长利, 赵越, 等. 高光谱成像技术的玉米叶片氮含量检测模型. 农机化研究, 2017, 39(11):140-147.
[1] Wang Yandan, Gao Xin, Peng Jinjian, Tang Feiyu. Comparison of Carbohydrate and Nitrogen Contents in Vegetative Organs between Early- and Middle-Maturing Cotton Lines and the Relationships to Dry Matter Accumulation [J]. Crops, 2023, 39(2): 106-114.
[2] Tang Zhongjie, Xie Deyi, Xu Shouming, Nie Lihong, Lü Shuping, Wang Mingkun. Changes of Insect Resistance and Its Correlation Analysis with Yield Traits in Transgenic Bt Cotton from 2005 to 2020 [J]. Crops, 2023, 39(2): 77-82.
[3] Li Wenshan, Zhang Junyao, Tang Jianghua, Xu Wenxiu, Xu Qinghua. Effects of Different Doses of AFD on Growth and Yield of Cotton [J]. Crops, 2023, 39(1): 158-162.
[4] Zhang Haitao, Yan Xiaobin, Feng Meichen, Yang Wude, Xiao Lujie, Song Xiaoyan, Zhang Meijun, Wang Lele, Wang Chao. Hyperspectral Response Characteristics and Monitor on Soil Water [J]. Crops, 2023, 39(1): 233-238.
[5] Ma Chunmei, Tian Yangqing, Zhao Qiang, Li Jiangyu, Wu Xueqin. Effects of Plant Growth Regulator Compound on Cotton Yield [J]. Crops, 2022, 38(6): 181-185.
[6] Feng Changhui, Jiao Chunhai, Zhang Youchang, Bie Shu, Qin Hongde, Wang Qiongshan, Zhang Jiaohai, Wang Xiaogang, Xia Songbo, Lan Jiayang, Chen Quanqiu. Genetic Analysis for Yield and Fiber Quality Traits in Upland Cotton Based on Partial NCII Mating Design [J]. Crops, 2022, 38(5): 13-21.
[7] Xu Min, Jin Lulu, Li Ruichun, Sun Liyuan, Wang Zisheng. Study on Cotton Chemical Topping in Liaohe Cotton Area [J]. Crops, 2022, 38(5): 201-207.
[8] Zhao Kang, Yang Tao, Wang Honggang, Li Shengmei, Pang Bo, Ma Shangjie, Gao Wenwei. Evaluation on Salt Tolerance of 42 Sea-Island Cotton (Gossypium barbadense) Varieties in Xinjiang during Germination Period [J]. Crops, 2022, 38(5): 27-33.
[9] Zhang Te, Li Guangwei, Li Kexin, Li Xinxin, Zhao Qiang. Effects of DPC through Drip Irrigation on Growth and Yield of Cotton [J]. Crops, 2022, 38(4): 124-131.
[10] Zhang Wen, Liu Quanyi, Zeng Qingtao, Cai Xiaoli, Feng Yang, Lu Tao. Effects of Different Row Spacings on Boll Characteristics and Fiber Quality of Machine Picked Cotton [J]. Crops, 2021, 37(2): 147-152.
[11] Sun Zhengran, Wu Hao, Zhang Cuiping, Zhang Jinli, He Daohua. Developing and Screening of Chemical Regulator for Cotton Topping [J]. Crops, 2021, 37(1): 112-117.
[12] Zhou Jiang, Xie Yizhang, Xiang Ping'an. Emergy Analysis of Inputs and Outputs of Major Field Crop Ecosystems in Hunan Province [J]. Crops, 2021, 37(1): 175-181.
[13] Ji Jingchun, Liu Jianli, Niu Yujie, Xuan Kefan, Jiang Yifei, Deng Haodong, Li Xiaopeng. Comparison of Estimation Methods for Growth Parameters of Winter Wheat Based on Full-Band Hyperspectral Data [J]. Crops, 2020, 36(6): 180-188.
[14] Qin Hongde, Rong Yihua, Huang Xiaoli, Hu Aibing, Zhou Jiahua, Yan Xianhui, Li Wei, Zhang Xianhong, Li Hongju, Yang Guozheng. Responses of Cotton to Planting Densities and Nitrogen Rates under Direct Seeding in Summer with Simplified Fertilization [J]. Crops, 2020, 36(4): 127-134.
[15] Zhang Qian, Li Yaofa, Wang Shulin, Wang Yan, Feng Guoyi, Lin Yongzeng, Liang Qinglong, Lei Xiaopeng, Qi Hong. Effects of Strip-Planting of Cotton-Wheat on Cotton Aphid [J]. Crops, 2020, 36(4): 206-210.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!