Modeling Method of Cotton Leaves SPAD at Flowering and Boll Stage in North China Plain Based on UAV Multi-Spectrum

doi:10.11924/j.issn.1000-6850.casb2020-0683

Abstract

Abstract:

SPAD spectral characteristics of cotton leaves in North China Plain need to be ascertained, and the most suitable modeling method is yet to be studied. Aiming at the cotton area of North China Plain and based on unmanned drone multi-spectrum, this paper explored SPAD spectral characteristics of cotton leaves and the best modeling method. Focusing on the cotton area of North China Plain in the Yellow River Basin, we took Dali village cotton area of Xiajin County, Dezhou City as research area, used unmanned drone to obtain multispectral images at flowering and boll stage, and simultaneously determined SPAD value of cotton leaves. In this paper, the original spectrum was preprocessed and combined to construct the spectral index, and then 6 cotton SPAD characteristic spectral indexes were screened out by correlation analysis. BP neural network (BPNN), multiple stepwise regression (MSR) and support vector machine (SVM) methods were used respectively to construct quantitative analysis model of SPAD value of cotton, and verified, compared and optimized the best model and modeling method, and then quantitatively analyzed the spatial distribution of cotton leaf SPAD in the research area. The results show that: the characteristic bands of cotton leaf SPAD are red band and red edge band. The characteristic spectral index of the selected model are r, r*reg, (reg-r)/(reg+r), r-g, r/g and. $\sqrt{r^{2}+g^{2}}$ Compared the three modeling methods, BPNN model has the highest accuracy. Its modeling set R² and RMSE are 0.747 and 4.568, respectively, and its verification set R², RMSE and RPD are 0.758, 4.142 and 2.135, respectively, and the model is determined as the best one of cotton leaf SPAD. Based on BP neural network model, the spatial distribution of cotton leaf SPAD is inverted, and the inversion value is highly consistent with the measured value, and the fitting result is good. BP neural network could be used as a preferred method for SPAD modeling of cotton leaves of North China Plain based on unmanned drone multi-spectrum. This research could promote quantitative remote sensing of cotton field and monitoring of cotton growth.

Key words: cotton, SPAD, UAV multi-spectrum, remote sensing, North China Plain

CLC Number:

S127

Ji Weishuai, Chen Hongyan, Wang Shuting, Zhang Yuting. Modeling Method of Cotton Leaves SPAD at Flowering and Boll Stage in North China Plain Based on UAV Multi-Spectrum[J]. Chinese Agricultural Science Bulletin, 2021, 37(22): 143-150.

Figures/Tables 11

波段	相关系数
红光(red)	-0.794^**
红边(reg)	0.496^*
绿光(green)	0.185
近红(nir)	0.149

序号	光谱指数	相关系数	序号	光谱指数	相关系数	序号	光谱指数	相关系数
1	$g + r$	0.7252^**	10	$g * r * reg$	-0.5523^**	20	$g 2 + re g 2$	0.3071^*
2	$r * nir$	-0.7124^**	11	$g + reg$	0.3197^*	21	$r + reg$	-0.7091^**
3	$r * reg$	-0.7279^**	12	$g * reg$	0.3027^*	22	$r * g$	-0.7059^**
4	$r + reg$	-0.1359	13	$(reg - g) / (reg + g)$	0.1231	23	$reg * g$	0.3068^*
5	$(reg - r) / (reg + r)$	0.7645^**	14	$r - g$	-0.81^**	24	$(reg + g) / (g - reg)$	0.0796
6	$g *nir$	0.1457	15	$reg - r$	0.5745	25	$g 2 + re g 2 + r 2$	0.1323
7	$g nir r$	-0.5937^*	16	$reg / r$	0.7091^**	26	$g * reg * r$	-0.5311^**
8	$g + r + reg$	-0.0966	17	$r / g$	-0.7995^**	27	$nir + g + reg$	0.1808^*
9	$g*r$	-0.6991^**	18	$r 2 + g 2$	-0.7254^**
10	$grreg$	-0.5523^**	19	$r 2 + re g 2$	0.1249

序号	光谱指数	相关系数	序号	光谱指数	相关系数	序号	光谱指数	相关系数
1	$g + r$	0.7252^**	10	$g * r * reg$	-0.5523^**	20	$g 2 + re g 2$	0.3071^*
2	$r * nir$	-0.7124^**	11	$g + reg$	0.3197^*	21	$r + reg$	-0.7091^**
3	$r * reg$	-0.7279^**	12	$g * reg$	0.3027^*	22	$r * g$	-0.7059^**
4	$r + reg$	-0.1359	13	$(reg - g) / (reg + g)$	0.1231	23	$reg * g$	0.3068^*
5	$(reg - r) / (reg + r)$	0.7645^**	14	$r - g$	-0.81^**	24	$(reg + g) / (g - reg)$	0.0796
6	$g *nir$	0.1457	15	$reg - r$	0.5745	25	$g 2 + re g 2 + r 2$	0.1323
7	$g nir r$	-0.5937^*	16	$reg / r$	0.7091^**	26	$g * reg * r$	-0.5311^**
8	$g + r + reg$	-0.0966	17	$r / g$	-0.7995^**	27	$nir + g + reg$	0.1808^*
9	$g*r$	-0.6991^**	18	$r 2 + g 2$	-0.7254^**
10	$grreg$	-0.5523^**	19	$r 2 + re g 2$	0.1249

建模方法	建模精度		验证精度
建模方法	决定系数R²	均方根误差RMSE	决定系数R²	均方根误差RMSE	相对分析误差RPD
多元逐步回归	0.682	4.667	0.709	4.487	1.791
支持向量机	0.709	4.974	0.695	4.850	1.878
BP神经网络	0.747	4.568	0.758	4.142	2.135

等级	反演值		实测值
等级	个数/个	所占比例/%	个数/个	所占比例/%
<28	3	3.19	5	5.32
28~34	7	7.45	7	7.45
34~36	4	4.26	3	3.19
36~38	4	4.26	4	4.26
38~40	2	2.13	4	4.26
40~42	13	13.83	12	12.77
42~44	17	18.09	14	14.89
44~46	13	13.83	10	10.64
46~48	13	13.83	11	11.70
48~50	5	5.32	7	7.45
50~52	5	5.32	6	6.38
>52	8	8.51	11	11.70

References 33

[1]	秦其明, 范闻捷, 任华忠, 等. 农田定量遥感理论、方法与应用[M]. 北京: 科学出版社, 2018.
[2]	吴炳方, 张淼, 曾红伟, 等. 大数据时代的农情监测与预警[J]. 遥感学报, 2016, 20(5):1027-1037.
[3]	Setiyono T D, Quicho E D, Gatti L, et al. Spatial rice yield estimation based on MODIS and sentinel-1 SAR data and ORYZA crop growth model[J]. Remote Sensing, 2018, 10(2):293. doi: 10.3390/rs10020293 URL
[4]	Ballesteros R, Ortega J F, Hernandez D, et al. Onion biomass monitoring using UAV-based RGB imaging[J]. Precision Agriculture, 2018, 19(5):840-857. doi: 10.1007/s11119-018-9560-y URL
[5]	毛智慧, 邓磊, 孙杰, 等. 无人机多光谱遥感在玉米冠层叶绿素预测中的应用研究[J]. 光谱学与光谱分析, 2018, 38(9):2923-2931.
[6]	Guillot B, Pouget F. UAV application incoastal environment, example of the oleron island for dunes and dikes survey[J]. ISPRS-International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2015, XL-3/W3(1):321-326. doi: 10.5194/isprsarchives-XL-3-W3-321-2015 URL
[7]	孙刚, 黄文江, 陈鹏飞, 等. 轻小型无人机多光谱遥感技术应用进展[J]. 农业机械学报, 2018, 49(3):1-17.
[8]	孙中宇, 陈燕乔, 杨龙, 等. 轻小型无人机低空遥感及其在生态学中的应用进展[J]. 应用生态学报, 2017, 28(2):528-536.
[9]	胡建波, 张建. 无人机遥感在生态学中的应用进展[J]. 生态学报, 2018, 38(1):20-30.
[10]	陈燕, 王登伟, 黄春燕, 等. 新疆棉花LAI和叶绿素密度的高光谱估算研究[J]. 遥感信息, 2007(2):33-36,41.
[11]	依尔夏提·阿不来提, 白灯莎·买买提艾力, 买买提·沙吾提, 等. 基于高光谱和BP神经网络的棉花冠层叶绿素含量联合估算[J]. 光学学报, 2019, 39(9):372-380.
[12]	陈硕博. 无人机多光谱遥感反演棉花光合参数与水分的模型研究[D]. 杨凌:西北农林科技大学, 2019.
[13]	罗丹, 常庆瑞, 齐雁冰, 等. 基于光谱指数的冬小麦冠层叶绿素含量估算模型研究[J]. 麦类作物学报, 2016, 36(9):1225-1233.
[14]	齐雁冰, 楚万林, 解飞, 等. 基于高光谱的渭北旱塬区棉花冠层叶面积指数估算[J]. 干旱地区农业研究, 2017, 35(1):114-121.
[15]	田明璐, 班松涛, 常庆瑞, 等. 基于无人机成像光谱仪数据的棉花叶绿素含量反演[J]. 农业机械学报, 2016, 47(11):285-293.
[16]	于雷, 洪永胜, 耿雷, 等. 基于偏最小二乘回归的土壤有机质含量高光谱估算[J]. 农业工程学报, 2015, 31(14):103-109.
[17]	王丹阳, 陈红艳, 王桂峰, 等. 无人机多光谱反演黄河口重度盐渍土盐分的研究[J]. 中国农业科学, 2019, 52(10):1698-1709.
[18]	周晓敏, 赵力彬, 张新利. 低空无人机影像处理技术及方法探讨[J]. 测绘与空间地理信息, 2012, 35(2):182-184.
[19]	奚雪, 赵庚星. 基于无人机多光谱遥感的冬小麦叶绿素含量反演及监测[J]. 中国农学通报, 2020, 36(20):119-126.
[20]	厉彦玲, 赵庚星, 常春艳, 等. OLI与HSI影像融合的土壤盐分反演模型[J]. 农业工程学报, 2017, 33(21):173-180.
[21]	刘明杰, 徐卓揆, 郜允兵, 等. 基于机器学习的稀疏样本下的土壤有机质估算方法[J]. 地球信息科学学报, 2020, 22(9):1799-1813. doi: 10.12082/dqxxkx.2020.190441
[22]	张卓然, 常庆瑞, 张廷龙, 等. 基于支持向量机的棉花冠层叶片叶绿素含量高光谱遥感估算[J]. 西北农林科技大学学报:自然科学版, 2018, 46(11):39-45.
[23]	李媛媛, 常庆瑞, 刘秀英, 等. 基于高光谱和BP神经网络的玉米叶片SPAD值遥感估算[J]. 农业工程学报, 2016, 32(16):135-142.
[24]	Senthilnath J, Manasa K, Akanksha D, et al. Application of UAV imaging platform for vegetation analysis based on spectral-spatial methods[J]. Computers and Electronics in Agriculture, 2017, 140:8-24 doi: 10.1016/j.compag.2017.05.027 URL
[25]	雷亚平, 韩迎春, 王国平, 等. 无人机低空数字图像诊断棉花苗情技术[J]. 中国棉花, 2017, 44(5):23-25.
[26]	李静, 王建军, 朱安. 基于低成本无人机的水稻叶片SPAD值遥感估测[J]. 吉林农业, 2017(18):68.
[27]	陈红艳, 赵庚星, 陈敬春, 等. 基于改进植被指数的黄河口区盐渍土盐分遥感反演[J]. 农业工程学报, 2015, 31(5):107-114.
[28]	梁后军, 郭蓉榕, 谢睿, 等. 基于神经网络的棉花产量预测[J]. 中国纤检, 2020(6):126-128.
[29]	范迎迎, 钱育蓉, 杨柳, 等. 基于BP神经网络的遥感影像棉花识别方法[J]. 计算机工程与设计, 2017, 38(5):1356-1360.
[30]	Urselmans T T, Schmidt H, Joergensen R G, et al. Usefulness of near-infrared spectroscopy to determine biological and chemical soil properties: Importance of sample pre-treatment[J]. Soil Biology and Biochemistry, 2008, 40(5):1178-1188. doi: 10.1016/j.soilbio.2007.12.011 URL
[31]	孙勃岩, 常庆瑞, 刘梦云. 冬小麦冠层叶绿素质量分数高光谱遥感反演研究[J]. 西北农业学报, 2017, 26(4):552-559.
[32]	杨可明, 张婉婉, 卓伟, 等. 红边光谱谐波分析的神经网络法叶绿素含量反演研究[J]. 科学技术与工程, 2016, 16(24):19-24.
[33]	刘文雅, 潘洁. 基于神经网络的马尾松叶绿素含量高光谱估算模型[J]. 应用生态学报, 2017, 28(4):1128-1136.