基于直方图的手机玉米冠层数字图像氮素诊断方法

doi:10.3864/j.issn.0578-1752.2024.20.014

摘要/Abstract

摘要：

【目的】便捷准确地诊断作物氮素营养状况是实现作物精准施肥和氮肥资源合理利用的关键。近年来应用数码相机等工具进行作物营养诊断的研究受到广泛关注。本研究采用智能手机相机获取玉米冠层图像，建立完善的基于手机相机的氮素营养诊断技术，并比较传统的图像均值方法和直方图方法对氮素营养诊断的可靠性，以探明夏玉米氮素营养诊断的最佳适用模型。【方法】基于田间氮肥用量试验，采用智能手机相机获取夏玉米拔节期冠层图像，提取夏玉米冠层图像的G/R、G/B、NRI [R/(R+G+B)]、NGI [G/(R+G+B)]、NBI [B/(R+G+B)]和(G-R)/(R+G+B) 6种颜色指数均值及直方图敏感区间，分别建立冠层图像色彩参数均值模型与直方图模型，分析其与玉米叶片含氮量和产量的关系。利用决定系数（R²）、均方根误差（RMSE）、平均绝对百分比误差（MAPE）对比不同指数模型模拟估算玉米叶片含氮量和产量的稳定性和准确性，建立基于手机相机获取夏玉米冠层图像的氮素营养诊断模型。【结果】施氮量显著影响玉米叶片含氮量、产量及冠层图像色调和植被覆盖度。直方图波峰b随叶片含氮量的增加而发生变化，相较于冠层图像色彩参数指数均值方法，指数直方图法适用于不同品种的氮素诊断。色彩参数(G-R)/(R+G+B)直方图可以更好地反映作物覆盖率及整体颜色信息，指数直方图与玉米叶片含氮量和产量也呈现较好的相关性。基于神经网络模型验证数据集精度评价指标，指数直方图模型中玉米叶片含氮量和产量的MAPE值和RMSE值均低于指数均值模型，R²达到0.753，大于指数均值模型。指数直方图模型验证结果MAPE值达到5.80%，RMSE值为0.07，估算精度高，泛化性强。结果表明，冠层图像色彩参数指数直方图在估算叶片含氮量和产量时具有更高精度和更强鲁棒性，能够有效利用玉米叶片覆盖度、颜色等特点，具有较好的稳定性。【结论】bbbbb利用智能手机相机获取玉米冠层数字图像，结合冠层图像色彩参数直方图方法建立的神经网络模型具有较好的应用效果，提高了估测精度，作为一种新方法在玉米氮素营养快速无损诊断和精准施肥中具有较好的应用潜力。

关键词: 氮素营养诊断, 手机相机, 夏玉米, 冠层, 神经网络模型, 数字图像技术

Abstract:

【Objective】Convenient and accurate diagnosis of crop nitrogen (N) status is the key to achieve precise crop fertilization and rational utilization of N resources. In recent years, the application of digital cameras and other tools in crop nutrition diagnosis has attracted wide attention. In this study, the smart phone cameras were used to obtain maize canopy images, and nitrogen nutrition diagnosis technology based on mobile phone cameras was established and improved. The reliability of traditional image mean method and histogram method for nitrogen nutrition diagnosis was compared to find out the best model for nitrogen nutrition diagnosis of summer maize. 【Method】Based on the experiment of N fertilizer amount in the field, the canopy image of summer maize at jointing stage was obtained by smartphone camera. Six color indices, including G/R, G/B, NRI [R/(R+G+B)], NGI [G/(R+G+B)], NBI [B/(R+G+B)] and (G-R)/(R+G+B), were extracted from summer maize canopy images, and the histogram sensitive interval were established, respectively, to analyze their relationship with leaf N content and yield of maize. The determination coefficient (R²) and root mean square error (RMSE) were used to determine the relationship between the mean color index model and the histogram model. Mean absolute percentage error (MAPE) was used to simulate and estimate the stability and accuracy of leaf N content and yield in maize compared with different index models. Then, the N nutrition diagnosis model based on mobile phone camera acquisition of summer maize canopy images was established. 【Result】N application significantly affected leaf N content, yield, canopy image hue and vegetation coverage of maize. The peak b of the histogram changes with the increase of leaf N content. Compared with the mean color index method in canopy images, the index histogram method was suitable for N diagnosis among different varieties. The color index (G-R)/(R+G+B) histogram could better reflect crop coverage and overall color information. The index histogram also showed a good correlation with leaf N content and yield. Based on the neural network model to validate the accuracy evaluation indicators of the dataset, the MAPE and RMSE values of leaf N content and yield in maize in the exponential histogram model were lower than those in the exponential mean model, and the R² reached 0.753, which was greater than that in the exponential mean model. The validation results of the exponential histogram model showed a MAPE value of 5.80% and an RMSE value of 0.07, indicating high estimation accuracy and strong generalization ability. The results indicated that the color parameter index histogram of canopy images had higher accuracy and stronger robustness in estimating leaf N content and yield, and could effectively utilize the characteristics of maize leaf coverage, color, etc., with good stability.【Conclusion】Therefore, the neural network model established by using smartphones to obtain digital images of maize canopy and combining them with the color index histogram method of canopy images has good application effects and improves estimation accuracy. As a new method, it has good potential in rapid and non-destructive diagnosis of maize N nutrition and precise fertilization.

Key words: nitrogen nutrition diagnosis, mobile phone camera, summer maize, canopy, neural network model, digital image technology

齐欣, 汪洋, 黄玉芳, 叶优良, 郭宇龙, 赵亚南. 基于直方图的手机玉米冠层数字图像氮素诊断方法[J]. 中国农业科学, 2024, 57(20): 4094-4106.

QI Xin, WANG Yang, HUANG YuFang, YE YouLiang, GUO YuLong, ZHAO YaNan. Nitrogen Nutrition Diagnosis Method Based on Mobile Phone Image of Summer Maize Canopy[J]. Scientia Agricultura Sinica, 2024, 57(20): 4094-4106.

0
/ / 推荐

导出引用管理器 EndNote|Reference Manager|ProCite|BibTeX|RefWorks

链接本文: https://www.chinaagrisci.com/CN/10.3864/j.issn.0578-1752.2024.20.014

https://www.chinaagrisci.com/CN/Y2024/V57/I20/4094

图/表 10

图1

表1

表2

表3

图2

表4

图3

图4

图5

表5

参考文献 35

[1]	赵亚南, 宿敏敏, 吕阳, 况福虹, 陈轩敬, 张跃强, 石孝均. 减量施肥下小麦产量、肥料利用率和土壤养分平衡. 植物营养与肥料学报, 2017, 23(4): 864-873.
	ZHAO Y N, SU M M, LÜ Y, KUANG F H, CHEN X J, ZHANG Y Q, SHI X J. Wheat yield, nutrient use efficiencies and soil nutrient balance under reduced fertilizer rate. Journal of Plant Nutrition and Fertilizer, 2017, 23(4): 864-873. (in Chinese)
[2]	赵亚南, 徐霞, 黄玉芳, 孙笑梅, 叶优良. 河南省小麦、玉米氮肥需求及节氮潜力. 中国农业科学, 2018, 51(14): 2747-2757. doi: 10.3864/j.issn.0578-1752.2018.14.012.
	ZHAO Y N, XU X, HUANG Y F, SUN X M, YE Y L. Nitrogen requirement and saving potential for wheat and maize in Henan Province. Scientia Agricultura Sinica, 2018, 51(14): 2747-2757. doi: 10.3864/j.issn.0578-1752.2018.14.012. (in Chinese)
[3]	魏雪, 贾彪, 兰宇, 马胜利, 马健祯, 蒋鹏, 孙权. 智能手机图像参数与玉米氮素营养状况关联解析. 生态学杂志, 2021, 40(8): 2656-2664.
	WEI X, JIA B, LAN Y, MA S L, MA J Z, JIANG P, SUN Q. Correlation analysis between smartphone image parameters and nitrogen nutrition states of maize. Chinese Journal of Ecology, 2021, 40(8): 2656-2664. (in Chinese) doi: DOI: 10.13292/j.1000-4890.202108.034
[4]	石蒙蒙, 王雪峰, 袁莹, 陈飞飞, 黄川腾, 王鹏, 陈星京. 基于冠层图像的海南粗榧苗期生长状态无损监测方法研究. 植物营养与肥料学报, 2023, 29(11): 2181-2192.
	SHI M M, WANG X F, YUAN Y, CHEN F F, HUANG C T, WANG P, CHEN X J. Non-destructive monitoring of Cephalotaxus mannii growth status at seedling stage based on canopy images. Journal of Plant Nutrition and Fertilizers, 2023, 29(11): 2181-2192. (in Chinese)
[5]	郭松, 常庆瑞, 郑智康, 蒋丹垚, 高一帆, 宋子怡, 姜时雨. 基于无人机高光谱影像的玉米叶绿素含量估测. 江苏农业学报, 2022, 38(4): 976-984.
	GUO S, CHANG Q R, ZHENG Z K, JIANG D Y, GAO Y F, SONG Z Y, JIANG S Y. Estimation of maize chlorophyll content based on unmanned aerial vehicle (UAV) hyperspectral images. Jiangsu Journal of Agricultural Sciences, 2022, 38(4): 976-984. (in Chinese)
[6]	INTARAVANNE Y, SUMRIDDETCHKAJORN S. Android-based rice leaf color analyzer for estimating the needed amount of nitrogen fertilizer. Computers and Electronics in Agriculture, 2015, 116: 228-233.
[7]	凌启鸿, 王绍华, 丁艳锋, 李刚华. 关于用水稻“顶3顶4叶叶色差”作为高产群体叶色诊断统一指标的再论证. 中国农业科学, 2017, 50(24): 4705-4713. doi: 10.3864/j.issn.0578-1752.2017.24.004.
	LING Q H, WANG S H, DING Y F, LI G H. Re-evaluation of using the color difference between the top 3rd leaf and the 4th leaf as a unified indicator for high-yielding rice. Scientia Agricultura Sinica, 2017, 50(24): 4705-4713. doi: 10.3864/j.issn.0578-1752.2017.24.004. (in Chinese)
[8]	BURGOS-ARTIZZU X P, RIBEIRO A, TELLAECHE A, PAJARES G, FERNÁNDEZ-QUINTANILLA C. Analysis of natural images processing for the extraction of agricultural elements. Image and Vision Computing, 2010, 28(1): 138-149.
[9]	武改红, 冯美臣, 杨武德, 王超, 孙慧, 贾学勤, 张松, 乔星星. 冬小麦叶片SPAD值高光谱估测的预处理方法. 生态学杂志, 2018, 37(5): 1589-1594.
	WU G H, FENG M C, YANG W D, WANG C, SUN H, JIA X Q, ZHANG S, QIAO X X. Hyperspectral pretreatment methods on leaf SPAD value prediction in winter wheat. Chinese Journal of Ecology, 2018, 37(5): 1589-1594. (in Chinese)
[10]	周琼, 杨红云, 杨珺, 孙玉婷, 孙爱珍, 杨文姬. 基于BP神经网络和概率神经网络的水稻图像氮素营养诊断. 植物营养与肥料学报, 2019, 25(1): 134-141.
	ZHOU Q, YANG H Y, YANG J, SUN Y T, SUN A Z, YANG W J. Feasibility study of BP neural network and probabilistic neural network for nitrogen nutrition diagnosis of rice images. Journal of Plant Nutrition and Fertilizers, 2019, 25(1): 134-141. (in Chinese)
[11]	张立周, 王殿武, 张玉铭, 程一松, 李红军, 胡春胜. 数字图像技术在夏玉米氮素营养诊断中的应用. 中国生态农业学报, 2010, 18(6): 1340-1344.
	ZHANG L Z, WANG D W, ZHANG Y M, CHENG Y S, LI H J, HU C S. Diagnosis of N nutrient status of corn using digital image processing technique. Chinese Journal of Eco-Agriculture, 2010, 18(6): 1340-1344. (in Chinese)
[12]	贺英, 邓磊, 毛智慧, 孙杰. 基于数码相机的玉米冠层SPAD遥感估算. 中国农业科学, 2018, 51(15): 66-77. doi: 10.3864/j.issn.0578-1752.2018.15.005.
	HE Y, DENG L, MAO Z H, SUN J. Remote sensing estimation of canopy SPAD value for maize based on digital camera. Scientia Agricultura Sinica, 2018, 51(15): 66-77. doi: 10.3864/j.issn.0578-1752.2018.15.005. (in Chinese)
[13]	张玲, 陈新平, 贾良良. 基于无人机可见光遥感的夏玉米氮素营养动态诊断参数研究. 植物营养与肥料学报, 2018, 24(1): 261-269.
	ZHANG L, CHEN X P, JIA L L. Parameter research of using UAV-based visible spectral analysis technology in dynamical diagnosis of nitrogen status of summer maize. Journal of Plant Nutrition and Fertilizers, 2018, 24(1): 261-269. (in Chinese)
[14]	张珏, 田海清, 李哲, 李斐, 史树德. 基于数码相机图像的甜菜冠层氮素营养监测. 农业工程学报, 2018, 34(1): 157-163.
	ZHANG J, TIAN H Q, LI Z, LI F, SHI S D. Nitrogen nutrition monitoring of beet canopy based on digital camera image. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(1): 157-163. (in Chinese)
[15]	JIA B, HE H B, MA F Y, DIAO M, JIANG G Y, ZHENG Z, CUI J, FAN H. Use of a digital camera to monitor the growth and nitrogen status of cotton. The Scientific World Journal, 2014(6): 171-192
[16]	BENDIG J, BOLTEN A, BENNERTZ S, BROSCHEIT J, EICHFUSS S, BARETH G. Estimating biomass of barley using crop surface models (CSMs) derived from UAV-based RGB imaging. Remote Sensing, 2014, 6(11): 10395-10412.
[17]	翟勇全, 魏雪, 运彬媛, 马健祯, 贾彪. 基于无人机图像参数对滴灌条件下玉米氮素营养的动态诊断. 中国农业气象, 2022, 43(4): 308-320.
	ZHAI Y Q, WEI X, YUN B Y, MA J Z, JIA B. Dynamic diagnosis of nitrogen nutrition in maize under drip irrigation condition based on UAV image parameters. Chinese Journal of Agrometeorology, 2022, 43(4): 308-320. (in Chinese)
[18]	FERNANDEZ-GALLEGO J A, KEFAUVER S C, VATTER T, APARICIO GUTIÉRREZ N, NIETO-TALADRIZ M T, ARAUS J L. Low-cost assessment of grain yield in durum wheat using RGB images. European Journal of Agronomy, 2019, 105: 146-156.
[19]	ALI A M. Development of an algorithm for optimizing nitrogen fertilization in wheat using GreenSeeker proximal optical sensor. Experimental Agriculture, 2020, 56(5): 688-698.
[20]	陈奕山. 小农户在中国农业现代化进程中的作用及处境变化. 中国农业大学学报(社会科学版), 2021, 38(4): 19-30.
	CHEN Y S. The role of small farm household and the change of their situation in agricultural modernization process. Journal of China Agricultural University (Social Sciences), 2021, 38(4): 19-30. (in Chinese)
[21]	夏莎莎, 张聪, 李佳珍, 李红军, 张玉铭, 胡春胜. 基于手机相机获取玉米叶片数字图像的氮素营养诊断与推荐施肥研究. 中国生态农业学报, 2018, 26(5): 703-709.
	XIA S S, ZHANG C, LI J Z, LI H J, ZHANG Y M, HU C S. Diagnosis of nitrogen nutrient and recommended fertilization in summer corn using leaf digital images of cellphone camera. Chinese Journal of Eco-Agriculture, 2018, 26(5): 703-709. (in Chinese)
[22]	贾彪, 贺正. 基于手机图像反演的滴灌玉米光响应曲线特征参数研究. 农业机械学报, 2019, 50(7): 229-236.
	JIA B, HE Z. Inversion of light response curve characteristic parameters of maize based on cellphone images. Transactions of the Chinese Society for Agricultural Machinery, 2019, 50(7): 229-236. (in Chinese)
[23]	齐欣. 基于手机图像的冬小麦冠层数字图像氮素营养诊断模型[D]. 郑州: 河南农业大学, 2021.
	QI X. Nitrogen nutrition diagnosis model of winter wheat canopy digital image based on mobile phone image[D]. Zhengzhou: Henan Agricultural University, 2021. (in Chinese)
[24]	周琼. 基于机器视觉和高光谱的水稻氮素营养诊断方法研究[D]. 南昌: 江西农业大学, 2019.
	ZHOU Q. Study on diagnosis method of rice nitrogen nutrition based on machine vision and hyperspectral[D]. Nanchang: Jiangxi Agricultural University, 2019. (in Chinese)
[25]	于涧, 洪欣, 于泽翔, 马涛. 基于BP神经网络的江苏省粮食产量预测. 沈阳师范大学学报(自然科学版), 2023, 41(4): 316-320.
	YU J, HONG X, YU Z X, MA T. Prediction of grain yield in Jiangsu Province based on BP neural network. Journal of Shenyang Normal University (Natural Science Edition), 2023, 41(4): 316-320. (in Chinese)
[26]	运彬媛, 张昊, 翟勇全, 马健祯, 姬丽, 李稼润, 金学兰, 贾彪. 基于土壤氮素水平的玉米冠层SPAD值估算方法. 生态学杂志, 2024, 43(5): 1488-1497.
	YUN B Y, ZHANG H, ZHAI Y Q, MA J Z, JI L, LI J R, JIN X L, JIA B. Estimation method for canopy SPAD values of maize based on soil nitrogen level. Chinese Journal of Ecology, 2024, 43(5): 1488-1497. (in Chinese) doi: 10.13292/j.1000-4890.202405.005
[27]	李媛媛, 常庆瑞, 刘秀英, 严林, 罗丹, 王烁. 基于高光谱和BP神经网络的玉米叶片SPAD值遥感估算. 农业工程学报, 2016, 32(16): 135-142.
	LI Y Y, CHANG Q R, LIU X Y, YAN L, LUO D, WANG S. Estimation of maize leaf SPAD value based on hyperspectrum and BP neural network. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(16): 135-142. (in Chinese)
[28]	廉世勋, 李承志, 吴振国, 张华京, 毛向辉. 农用转光剂及转光膜开发进展. 中国塑料, 2000, 14(9): 1-5.
	LIAN S X, LI C Z, WU Z G, ZHANG H J, MAO X H. Progress in development of light conversion agents and films for agriculture. China Plastics, 2000, 14(9): 1-5. (in Chinese)
[29]	QI X, ZHAO Y N, HUANG Y F, WANG Y, QIN W, FU W, GUO Y L, YE Y L. A novel approach for nitrogen diagnosis of wheat canopies digital images by mobile phones based on histogram. Scientific Reports, 2021, 11: 13012. doi: 10.1038/s41598-021-92431-5 pmid: 34155294
[30]	ADAMSEN F J, PINTER P J Jr, BARNES E M, LAMORTE R L, WALL G W, LEAVITT S W, KIMBALL B A. Measuring wheat senescence with a digital camera. Crop Science, 1999, 39(3): 719-724.
[31]	李井会, 朱丽丽, 宋述尧. 数字图像技术在马铃薯氮素营养诊断中的应用. 中国马铃薯, 2006, 20(5): 257-260.
	LI J H, ZHU L L, SONG S Y. Diagnosis of N status of potato using digital image processing technique. Chinese Potato Journal, 2006, 20(5): 257-260. (in Chinese)
[32]	李红军, 张立周, 陈曦鸣, 张玉铭, 程一松, 胡春胜. 应用数字图像进行小麦氮素营养诊断中图像分析方法的研究. 中国生态农业学报, 2011, 19(1): 155-159.
	LI H J, ZHANG L Z, CHEN X M, ZHANG Y M, CHENG Y S, HU C S. Image analysis method in application of digital image on diagnosing wheat nitrogen status. Chinese Journal of Eco-Agriculture, 2011, 19(1): 155-159. (in Chinese)
[33]	贾良良. 应用数字图像技术与土壤植株测试进行冬小麦氮营养诊断[D]. 北京: 中国农业大学, 2003.
	JIA L L. Diagnosis of N status of winter wheat using digital image processing and soil-plant testing techniques[D]. Beijing: China Agricultural University, 2003. (in Chinese)
[34]	魏全全, 李岚涛, 任涛, 王振, 王少华, 李小坤, 丛日环, 鲁剑巍. 基于数字图像技术的冬油菜氮素营养诊断. 中国农业科学, 2015, 48(19): 3877-3886. doi: 10.3864/j.issn.0578-1752.2015.19.010.
	WEI Q Q, LI L T, REN T, WANG Z, WANG S H, LI X K, CONG R H, LU J W. Diagnosing nitrogen nutrition status of winter rapeseed via digital image processing technique. Scientia Agricultura Sinica, 2015, 48(19): 3877-3886. doi: 10.3864/j.issn.0578-1752.2015.19.010. (in Chinese)
[35]	WAN L, CEN H Y, ZHU J P, ZHANG J F, ZHU Y M, SUN D W, DU X Y, ZHAI L, WENG H Y, LI Y J, LI X R, BAO Y D, SHOU J Y, HE Y. Grain yield prediction of rice using multi-temporal UAV-based RGB and multispectral images and model transfer-a case study of small farmlands in the South of China. Agricultural and Forest Meteorology, 2020, 291: 108096.

品种 Cultivar	氮水平 Nitrogen level	叶片含氮量 Leaf N content (%)	产量 Yield (kg·hm^-2)
登海605 Denghai 605，DH605	N0	0.77±0.14b	5766±1817b
	N120	0.86±0.03b	8612±738a
	N180	1.07±0.04a	9469±772a
	N240	0.98±0.12ab	8499±522a
	N360	0.86±0.15b	6297±591b
	平均Mean	0.91	7729
	CV (%)	12.24	12.76
科玉188 Keyu 188，KY188	N0	0.57±0.02b	7202±1642a
	N120	0.89±0.08a	8131±200a
	N180	0.92±0.07a	8956±1418a
	N240	0.84±0.10a	8592±1022a
	N360	0.97±0.11a	9037±987a
	平均Mean	0.84	8384
	CV(%)	8.51	12.79

品种 Cultivar	氮水平 Nitrogen level	色彩参数 Color parameter
品种 Cultivar	氮水平 Nitrogen level	G/R	G/B	NRI	NGI	NBI	(G-R)/(R+G+B)
登海605 Denghai 605，DH605	N0	0.93±0.24a	1.55±0.26a	0.25±0.04a	0.31±0.07a	0.16±0.04a	0.07±0.02a
	N120	0.99±0.22a	1.61±0.22a	0.26±0.03a	0.33±0.06a	0.18±0.04a	0.07±0.03a
	N180	1.07±0.15a	1.59±0.36a	0.28±0.05a	0.36±0.06a	0.21±0.03a	0.07±0.01a
	N240	1.10±0.09a	1.66±0.12a	0.27±0.01a	0.36±0.02a	0.20±0.01a	0.09±0.01a
	N360	1.01±0.04a	1.60±0.02a	0.27±0.00a	0.34±0.01a	0.19±0.01a	0.07±0.01a
	平均Mean	0.15	0.20	0.03	0.04	0.03	0.02
	CV(%)	14.90	12.44	10.51	12.98	13.89	24.31
科玉188 Keyu 188，KY188	N0	0.97±0.03a	2.15±0.28a	0.32±0.03a	0.38±0.03a	0.18±0.02ab	0.05±0.02a
	N120	0.95±0.02a	1.54±0.34b	0.27±0.02bc	0.34±0.03a	0.20±0.02a	0.05±0.01a
	N180	0.93±0.12a	1.54±0.21b	0.25±0.02c	0.34±0.04a	0.21±0.03a	0.05±0.01a
	N240	0.89±0.30a	1.47±0.43b	0.19±0.04d	0.29±0.10a	0.13±0.08b	0.07±0.03a
	N360	0.97±0.02a	1.85±0.18ab	0.31±0.01ab	0.37±0.02a	0.20±0.01a	0.05±0.01a
	平均Mean	0.10	0.29	0.02	0.05	0.03	0.02
	CV(%)	10.68	17.50	9.94	14.073	18.902	28.69

品种 Cultivar	色彩参数 Color parameter	叶片含氮量 Leaf N content (%)	产量 Yield (kg·hm^-2)
登海605 Denghai 605， DH605	G/R	0.334	0.563
	G/B	0.166	0.406
	NRI	0.256	0.568
	NGI	0.301	0.560
	NBI	0.368	0.632
	(G-R)/(R+G+B)	0.299	0.392
科玉188 Keyu 188，KY188	G/R	0.159	-0.460
	G/B	-0.237	-0.610
	NRI	-0.193	-0.326
	NGI	0.082	-0.530
	NBI	0.417	-0.166
	(G-R)/(R+G+B)	-0.017	-0.244

品种 Cultivar	色彩参数 Color parameter	叶片含氮量 Leaf N content (%)	产量 Yield (kg·hm^-2)
登海605 Denghai 605， DH605	G/R	0.456	0.345
	G/B	-0.342	-0.416
	NRI	0.588	0.689
	NGI	-0.070	-0.031
	NBI	0.432	0.526
	(G-R)/(R+G+B)	-0.633	-0.588
科玉188 Keyu 188， KY188	G/R	0.493	0.660
	G/B	-0.121	-0.667
	NRI	0.796	0.250
	NGI	0.394	0.625
	NBI	0.244	-0.257
	(G-R)/(R+G+B)	-0.550	-0.590

模型 Model	指标 Indicators	训练数据集 Training dataset		验证数据集 Validation dataset		R²
模型 Model	指标 Indicators	MAPE (%)	RMSE	MAPE (%)	RMSE	R²
指数均值 Color parameter mean	叶片含氮量 Leaf N content	20.48	0.211	14.60	0.223	0.265
指数均值 Color parameter mean	产量 Yield	12.97	1362.953	15.56	1481.131	0.435
指数直方图 Color parameter histogram	叶片含氮量 Leaf N content	10.92	0.126	5.80	0.070	0.753
指数直方图 Color parameter histogram	产量 Yield	10.42	1177.432	9.20	918.267	0.653