基于多光谱成像和卷积神经网络的玉米作物营养状况识别方法研究

doi:10.12133/j.smartag.2020.2.1.202001-SA001

摘要/Abstract

摘要：

水肥一体化自动装备的使用能够有效提高水肥资源利用率，但需要在作业前获知作物的营养状况及水肥需求量，而通过人工手持测量仪器来获取这些信息，存在着时效性差和劳动强度大等缺点。针对以上问题，本研究以常见的作物玉米为研究对象，使用大疆精灵Ⅲ无人机携带RedEdge-M多光谱相机在田间上空采集玉米多光谱图像，同时使用YLS-D系列植株营养测定仪测量玉米植株的氮素和水分含量等营养信息，根据这些信息将采集的图像分为3个等级（每个等级共包含530幅五通道图像，其中480幅作为训练集，50幅作为验证集），提出了一种基于卷积神经网络的玉米作物营养状况识别方法。并基于TensorFlow深度学习框架搭建了ResNet18卷积神经网络模型，通过向模型输入彩色图像数据和五通道多光谱图像数据，分别训练出适合于彩色图像和多光谱图像的玉米植株营养状况等级识别模型。试验结果表明：训练后的模型能够识别玉米作物的彩色图像和多光谱图像，能够输出玉米的营养状况等级和GPS 信息，识别彩色图像模型在验证集的正确率为84.7%，识别多光谱图像模型在验证集的正确率为90.5%，模型训练平均时间为4.5h，五通道图像识别平均用时为3.56s。该识别方法可快速无损地获取玉米作物的营养状况，为有效提高水肥资源利用率提供了方法和依据。

关键词: 智慧农业, 卷积神经网络, 多光谱图像, 玉米作物, 营养状况识别

Abstract:

Excessive application of water and fertilizer not only causes resources serious waste of, but also causes serious environmental pollution. The implementation of precision irrigation and fertilization can effectively reduce nutrient loss and environmental pollution, save irrigation water and improve the utilization rate of water and fertilizer resources, which is one of the important ways to promote the sustainable development of agriculture. The use of the integrated water-fertilizer equipment can effectively improve the utilization rate of water-fertilizer resources, but it is necessary to know the nutritional status of crops and water-fertilizer demand before operation. To acquire the information by hand-held measuring instruments, there are some disadvantages, such as poor timeliness and high labor intensity. In response to the above problems, this study took the common corn crop as an example, used the DJI Phantom III drone to carry RedEdge-M multispectral camera to collect multispectral images of corn crops over the fields, and measured nitrogen and moisture content of corn plants by YLS-D series plant nutrition tester. Based on this information, the collected images were divided into 3 levels, each level contains 530 five channel images (2650 single channel images), including 480 five channel images (2400 single channel images) in the training set and 50 five channel images (250 single channel images) in the verification set, and a method of identifying the nutritional status of corn crops based on convolutional neural network was proposed. Based on the TensorFlow deep learning framework, ResNet18 convolution neural network model was constructed. By entering color image data and five-channel multispectral image data into the model, the nutritional status recognition model of corn plant suitable for color image and multispectral image was trained, and the experimental results showed that the trained model could be used to recognize the multispectral images of corn, and the nutritional status of corn, topdressing guidance and GPS information could be outputted, the correct rate of the recognition color image model in the verification set was 84.7%. The correct rate of identifying multispectral image model in the verification set was 90.5%, the average time of model training was 4.5h, and the average time of recognizing a five channel image is 3.56 seconds, which can detect the nutritional status of corn crops quickly and undamaged, and provides a theoretical and technical basis for the accuracy of the application of water fertilizer in intelligent agriculture.

Key words: intelligent agriculture, convolution neural network, multispectral image, corn crop, nutritional recognition method

中图分类号:

TP75

吴刚, 彭要奇, 周广奇, 李晓龙, 郑永军, 严海军. 基于多光谱成像和卷积神经网络的玉米作物营养状况识别方法研究[J]. 智慧农业（中英文）, 2020, 2(1): 111-120.

Wu Gang, Peng Yaoqi, Zhou Guangqi, Li Xiaolong, Zheng Yongjun, Yan Haijun. Recognition method for corn nutrient based on multispectral image and convolutional neural network[J]. Smart Agriculture, 2020, 2(1): 111-120.

参考文献

1	国家统计局. 国家统计局关于2019年粮食产量数据的公告[EB/OL]. [2019-12-07].
1	http://www.gov.cn/xinwen/2019-12/07/content_5459250.htm, 2019-12-07.
2	王林权，周春菊. 夏玉米水肥异区交替灌溉施肥的产量与环境效应研究进展[J]. 植物营养与肥料学报, 2017, 23(6): 1651-1658.
2	Wang L, Zhou C. Advances in researches of yield and environmental effect of alternate furrow irrigation with separated water and N fertilizer supply on summer maize[J]. Plant Nutrition and Fertilizer Science, 2017, 23(6): 1651-1658.
3	向友珍, 张富仓, Rashad Mohamed A, 等. 灌溉施肥对尼罗河三角洲玉米产量和水分利用率的影响[J]. 农业机械学报, 2015, 46(10): 116-126.
3	Xiang Y, Zhang F, Rashad M, et al. Effects of different irrigation and fertilization strategies on yield and water use efficiency of maize in nile delta[J]. Transactions of the CSAM, 2015, 46(10): 116-126.
4	Gevaert C M, Suomalainen J, Tang J, et al. Generation of spectral-temporal response surfaces by combining multispectral satellite and hyperspectral UAV imagery for precision agriculture applications[J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2017, 8(6): 3140-3146.
5	Morin M, Lawrence R, Repasky K S, et al. Agreement analysis and spatial sensitivity of multispectral and hyperspectral sensors in detecting vegetation stress at management scales[J]. Journal of Applied Remote Sensing, 2017, 11: 1-11.
6	张卫正. 基于视觉与图像的植物信息采集与处理技术研究[D]. 杭州: 浙江大学, 2016.
6	Zhang W. Study of the plant information acquisition and processing technology based on vision and image[D]. Hangzhou: Zhejiang University, 2016.
7	贾彪, 马富裕. 基于机器视觉的棉花氮素营养诊断系统设计与试验[J]. 农业机械学报, 2016, 47(03): 305-310.
7	Jia B, Ma F. Design and experiment of nitrogen nutrition diagnosis system of cotton based on machine vision[J]. Transactions of the CSAM, 2016, 47(3): 305-310.
8	Wang D, Zhang M, Li J, et al. Intelligent constellation diagram analyzer using convolutional neural network-based deep learning[J]. Optics Express, 2017, 25(15): 17150-17166.
9	刘小刚, 范诚, 李加念, 等. 基于卷积神经网络的草莓识别方法[J]. 农业机械学报, 2020, 51(2): 237-244.
9	Liu X, Fan C, Li J, et al. Identification method of strawberry based on convolutional neural network[J]. Transactions of the CSAM, 2020, 51(2): 237-244.
10	叶发茂, 董萌, 罗威, 等. 基于卷积神经网络和重排序的农业遥感图像检索[J]. 农业工程学报, 2019, 35(15): 138-145.
10	Ye F, Dong M, Luo W, et al. Agricultural remote sensing image retrieval based on convolutional neural network and reranking[J]. Transactions of the CSAE, 2019, 35(15): 138-145.
11	汪传建, 赵庆展, 马永建, 等. 基于卷积神经网络的无人机遥感农作物分类[J]. 农业机械学报, 2019, 50(11): 161-168.
11	Wang C, Zhao Q, Ma Y, et al. Crop Identification of drone remote sensing based on convolutional neural network[J]. Transactions of the CSAM, 2019, 50(11): 161-168.
12	Dyrmann M, Karstoft H, Midtiby H S. Plant species classification using deep convolutional neural network[J]. Biosystems Engineering, 2016, 151: 72-80.
13	Yao C, Zhang Y, Liu H. Application of convolutional neural network in classification of high resolution agricultural remote sensing images[J]. ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, XLII-2/W7: 989-992.
14	陈鹏飞, 梁飞. 基于低空无人机影像光谱和纹理特征的棉花氮素营养诊断研究[J]. 中国农业科学, 2019, 52(13): 2220-2229.
14	Chen P, Liang F. Cotton nitrogen nutrition diagnosis based on spectrum and texture feature of images from low altitude unmanned aerial vehicle[J]. Scientia Agricultura Sinica, 2019, 52(13): 2220-2229.
15	李俊霞, 杨俐苹. 不同品种玉米氮含量与叶片光谱反射率及 SPAD 值的相关性[J].中国土壤与肥料, 2015, (3): 34-39.
15	Li J, Yang L. The correlation of total nitrogen content with leaf spectral reflectance and SPAD values in different maize varieties[J]. Soil and Fertilizer Sciences in China, 2015, (3): 34-39.
16	张银杰. 基于叶片光谱分析的玉米氮素营养诊断研究[D]. 北京: 中国农业科学院, 2019.
16	Zhang Y. Research on diagnosis of nitrogen status in maize based on leaf spectral analysis [D]. Beijing: Chinese Academy of Agricultural Sciences, 2019.
17	Sun Y, Xue B, Zhang M, et al. Completely automated CNN architecture design based on blocks[J]. IEEE Transactions on Neural Networks and Learning Systems, 2019, 31(4):1242-1254.
18	Dominguez-Sanchez A, Cazorla M, Orts-Escolano S. Pedestrian movement direction recognition using convolutional neural networks[J]. IEEE Transactions on Intelligent Transportation Systems, 2017, 18(12)： 3540-3548.
19	Wang Z, Hu M, Zhai G. Application of deep learning architectures for accurate and rapid detection of internal mechanical damage of blueberry using hyperspectral transmittance data[J]. Sensors (Basel), 2018, 18(4): 1126-1140.

[1]	潘晨露, 张正华, 桂文豪, 马家俊, 严晨曦, 张晓敏. 融合ECA机制与DenseNet201的水稻病虫害识别方法[J]. 智慧农业(中英文), 2023, 5(2): 45-55.
[2]	郭大方, 杜岳峰, 武秀恒, 侯思余, 栗晓宇, 张延安, 陈度. 农机装备数字孪生：从概念到应用[J]. 智慧农业(中英文), 2023, 5(2): 149-160.
[3]	李莉, 李民赞, 刘刚, 张漫, 汪懋华. 中国大田作物智慧种植目标、关键技术与区域模式[J]. 智慧农业(中英文), 2022, 4(4): 26-34.
[4]	胡瑞法, 刘万嘉文. 科技革命、颠覆性技术与智慧农业[J]. 智慧农业(中英文), 2022, 4(4): 138-143.
[5]	邵明月, 张建华, 冯全, 柴秀娟, 张凝, 张文蓉. 深度学习在植物叶部病害检测与识别的研究进展[J]. 智慧农业(中英文), 2022, 4(1): 29-46.
[6]	李少波, 杨玲, 于辉辉, 陈英义. 水下鱼类品种识别模型与实时识别系统[J]. 智慧农业(中英文), 2022, 4(1): 130-139.
[7]	周巧黎, 马丽, 曹丽英, 于合龙. 基于改进轻量级卷积神经网络MobileNetV3的番茄叶片病害识别[J]. 智慧农业(中英文), 2022, 4(1): 47-56.
[8]	王莹, 李越, 武婷婷, 孙石, 王敏娟. 基于密度估计和VGG-Two的大豆籽粒快速计数方法[J]. 智慧农业(中英文), 2021, 3(4): 111-122.
[9]	邱文杰, 叶进, 胡亮青, 杨娟, 李其利, 莫贱友, 易万茂. 面向植物病害识别的卷积神经网络精简结构Distilled-MobileNet模型[J]. 智慧农业(中英文), 2021, 3(1): 109-117.
[10]	洪葳, 胥保华, 刘升平. 蜂群多特征长期监测系统设计与试验研究[J]. 智慧农业（中英文）, 2020, 2(2): 105-114.
[11]	王培龙, 唐智勇. 农产品质量安全纳米传感应用研究分析与展望[J]. 智慧农业（中英文）, 2020, 2(2): 1-10.
[12]	杨成海. 机载遥感系统在精准农业中的应用[J]. 智慧农业（中英文）, 2020, 2(1): 1-22.
[13]	曹宏鑫, 葛道阔, 张文宇, 张伟欣, 曹静, 梁万杰, 宣守丽, 刘岩, 吴茜, 孙传亮, 张玲玲, 夏吉安, 刘永霞, 陈昱利, 岳延滨, 张智优, 万倩, 潘月, 韩旭杰, 吴菲. 农业模型发展分析及应用案例[J]. 智慧农业（中英文）, 2020, 2(1): 147-162.
[14]	万亮, 岑海燕, 朱姜蓬, 张佳菲, 杜晓月, 何勇. 基于纹理特征与植被指数融合的水稻含水量无人机遥感监测[J]. 智慧农业（中英文）, 2020, 2(1): 58-67.
[15]	吴才聪, 方向明. 基于北斗系统的大田智慧农业精准服务体系构建[J]. 智慧农业, 2019, 1(4): 83-90.