Variety identification model for maize seeds using hyperspectral pixel-level information combined with convolutional neural network

Liguo WANG; Lifeng WANG

doi:10.11834/jrs.20219349

Remote Sensing Meets Artificial Intelligence | Views : 0 下载量: 342 CSCD: 6 更多指标

PDF
Export
Share
Collection
Album

Variety identification model for maize seeds using hyperspectral pixel-level information combined with convolutional neural network
Vol. 25, Issue 11, Pages: 2234-2244(2021)
Published： 07 November 2021 ，
DOI： 10.11834/jrs.20219349

扫描看全文

王立国，王丽凤.2021.结合高光谱像素级信息和CNN的玉米种子品种识别模型.遥感学报，25（11）： 2234-2244

Wang L G and Wang L F. 2021. Variety identification model for maize seeds using hyperspectral pixel-level information combined with convolutional neural network. National Remote Sensing Bulletin， 25（11）：2234-2244
王立国，王丽凤.2021.结合高光谱像素级信息和CNN的玉米种子品种识别模型.遥感学报，25（11）： 2234-2244 DOI： 10.11834/jrs.20219349.

Wang L G and Wang L F. 2021. Variety identification model for maize seeds using hyperspectral pixel-level information combined with convolutional neural network. National Remote Sensing Bulletin， 25（11）：2234-2244 DOI： 10.11834/jrs.20219349.

摘要

玉米作为中国重要粮食作物，品种众多，易出现错分现象，影响农业安全和粮食生产。针对传统基于卷积神经网络CNN（Convolutional Neural Network）的高光谱图像作物品种识别模型所需建模样本数量巨大的问题，提出基于高光谱像素级信息和CNN的玉米种子品种识别模型。首先，获取不同品种玉米种子在400—1000 nm范围内的高光谱图像，提取样本全部像素的203维光谱信息，利用主成分分析PCA（Principal Component Analysis）算法将光谱维度降至8维。在实验中，样本的像素级光谱信息（即：样本的全部像素的光谱信息）除应用于CNN模型外，也应用于支持向量机（SVM）和K近邻分类（KNN）模型中，结果表明：在相同模型中，基于像素级光谱信息比基于米粒级光谱信息（即：每粒样本所有像素光谱信息的平均值）识别效果好；在相同情况下，CNN模型比SVM和KNN模型的识别效果好；基于像素级光谱信息和CNN的品种识别模型识别效果最稳定，依据像素级分类结果采用多数投票策略对玉米种子样本进行识别，样本识别精度高达100%（注：100%为建模集样本与测试集样本数量为0.27和0.32时的识别精度，随着测试集样本数量的增加，该识别精度将有所降低）。最后，使用t分布随机邻域嵌入（

-SNE）算法实现CNN输出特征值的可视化，验证了基于高光谱像素级信息和CNN的品种识别模型的有效性。在建模样本极少的情况下，实现了玉米种子品种的无损、高效识别，为精准农业提供了理论基础。

Abstract

As an important grain crop in China

maize has many varieties and is prone to misclassification

affecting agricultural security and food production. With the development of hyperspectral imaging and deep learning technology

identifying crop varieties using a combination of both is possible. A Convolutional Neural Network (CNN)

the most representative algorithm of deep learning technology to deal with the image classification task

needs a large number of training samples in the model training process. However

obtaining a large number of hyperspectral images of maize seed samples is difficult and time consuming. Aiming at the problem of the large number of modeling samples required for the traditional method based on CNN for crop identification in hyperspectral images

a variety identification model for maize seeds based on hyperspectral pixel-level spectral information and CNN is proposed. First

hyperspectral images of different varieties of maize seeds in the range of 400—1000 nm are obtained

and 203-dimensional spectral information of all the pixels of samples is extracted. Nevertheless

the enormous amount of spectral information creates the problem of dimensional disaster and greatly increases the computational cost. Second

to reduce the dimensionality of the sample spectral information

the principal component analysis algorithm is used to reduce the spectral dimension to eight dimensions

which effectively shortens the operation time. Third

the pixel-level spectral information of the sample (i.e.

the spectral information of all the pixels of the sample) is applied to the Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) classification models

in addition to the CNN model. The experimental results demonstrate that for the CNN

SVM

and KNN recognition algorithms

the pixel-level spectral information models show a more stable and efficient recognition effect than the seed-level one (i.e.

the average of all pixel spectral information of each sample). The seed-level information model does not fully utilize the sample pixel spectrum and spatial information

which needs a large number of modeling samples. When the number of samples used to build the classification model is the same

the CNN model has a significantly better recognition effect than the SVM and KNN models. In accordance with all the pixel-level classification results

a majority voting strategy is used to identify corn seed sample variety

and the sample recognition accuracy is up to 100% (100% refers to the identification accuracy when the numbers of samples in the modeling and test sets are 0.27 and 0.32

respectively. As the number of samples in the test set increases

the identification accuracy decreases). Lastly

the t-distribution random neighbor embedding algorithm is used to realize the visualization of output eigenvalues of CNN. The features of different maize seed varieties are clearly bounded in the visualization

which adequately verifies the validity of the species recognition model based on hyperspectral pixel-level information and CNN. In the rare case of modeling seed samples

nondestructive and efficient variety identification of maize seeds is realized

which will provide a theoretical basis for precision agriculture.

关键词

高光谱图像卷积神经网络深度学习玉米种子t分布随机邻域嵌入算法像素级光谱信息

Keywords

hyperspectral imageconvolutional neural networkdeep learningmaize seedt-distribution random neighbor embedding algorithmpixel-level spectral information

references

Alishahi A, Farahmand H, Prieto N and Cozzolino D. 2010. Identification of transgenic foods using NIR spectroscopy: a review. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 75(1): 1-7 [DOI: 10.1016/j.saa.2009.10.001http://dx.doi.org/10.1016/j.saa.2009.10.001]

Cao X L, Deng M J and Cui G X. 2019. Identifying ramie variety by combining the hyperspectral technology with the principal component analysis. Spectroscopy and Spectral Analysis, 39(6): 1905-1908

曹晓兰, 邓梦洁, 崔国贤. 2019. 高光谱结合主成分分析的苎麻品种识别. 光谱学与光谱分析, 39(6): 1905-1908) [DOI: 10.3964/j.issn.1000-0593(201906-1905-04]

Jia S Q. 2015. Study on the Purity Identification of Maize Seed Based on Near Infrared Imaging. Beijing: China Agricultural University: 1-5

贾仕强. 2015. 基于近红外光谱图像的玉米种子纯度鉴定方法研究. 北京: 中国农业大学: 1-5

Jolliffe I T. 1986. Principal Component Analysis. New York: Springer-Verlag: 111-130

LeCun Y, Bengio Y and Hinton G. 2015. Deep learning. Nature, 521(7553): 436-444 [DOI: 10.1038/nature14539http://dx.doi.org/10.1038/nature14539]

Li Q J，Liu J，Mi X F，Yang J and Yu T. 2021. Object-oriented crop classification for GF-6 WFV remote sensing images based on Convolutional Neural Network. National Remote Sensing Bulletin, 25(2): 549-558

李前景, 刘珺, 米晓飞, 杨健, 余涛. 2021. 面向对象与卷积神经网络模型的GF-6 WFV影像作物分类. 遥感学报, 25(02): 549-558［DOI：10.11834/jrs.20219347http://dx.doi.org/10.11834/jrs.20219347］

Maxwell A E, Warner T A and Fang F. 2018. Implementation of machine-learning classification in remote sensing: an applied review. International Journal of Remote Sensing, 39(9): 2784-2817 [DOI: 10.1080/01431161.2018.1433343http://dx.doi.org/10.1080/01431161.2018.1433343]

Nie P C, Zhang J N, Feng X P, Yu C L and He Y. 2019. Classification of hybrid seeds using near-infrared hyperspectral imaging technology combined with deep learning. Sensors and Actuators B: Chemical, 296: 126630 [DOI: 10.1016/j.snb.2019.126630http://dx.doi.org/10.1016/j.snb.2019.126630]

Paoletti M E, Haut J M, Plaza J and Plaza A. 2018. A new deep convolutional neural network for fast hyperspectral image classification. ISPRS Journal of Photogrammetry and Remote Sensing, 145: 120-147 [DOI: 10.1016/j.isprsjprs.2017.11.021http://dx.doi.org/10.1016/j.isprsjprs.2017.11.021]

Qiu Z J, Chen J, Zhao Y Y, Zhu S S, He Y and Zhang C. 2018. Variety identification of single rice seed using hyperspectral imaging combined with convolutional neural network. Applied Sciences, 8(2): 212 [DOI: 10.3390/app8020212http://dx.doi.org/10.3390/app8020212]

Rodríguez-Pulido F J, Barbin D F, Sun D W, Gordillo B, González-Miret M L and Heredia F J. 2013. Grape seed characterization by NIR hyperspectral imaging. Postharvest Biology and Technology, 76: 74-82 [DOI: 10.1016/j.postharvbio.2012.09.007http://dx.doi.org/10.1016/j.postharvbio.2012.09.007]

Scholkopf B and Smola A J. 2001. Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. Cambridge: MIT Press: 187-404

Serranti S, Cesare D, Marini F and Bonifazi G. 2013. Classification of oat and groat kernels using NIR hyperspectral imaging. Talanta, 103: 276-284 [DOI: 10.1016/j.talanta.2012.10.044http://dx.doi.org/10.1016/j.talanta.2012.10.044]

Shrestha S, Knapič M, Žibrat U, Deleuran L C and Gislum R. 2016. Single seed near-infrared hyperspectral imaging in determining tomato (Solanum lycopersicum L.) seed quality in association with multivariate data analysis. Sensors and Actuators B: Chemical, 237: 1027-1034 [DOI: 10.1016/j.snb.2016.08.170http://dx.doi.org/10.1016/j.snb.2016.08.170]

Van Der Maaten L and Hinton G. 2008. Visualizing data using t-SNE. Journal of Machine Learning Research, 9: 2579-2605

Wang L, Sun D W, Pu H B and Zhu Z W. 2016. Application of hyperspectral imaging to discriminate the variety of maize seeds. Food Analytical Methods, 9(1): 225-234 [DOI: 10.1007/s12161-015-0160-4http://dx.doi.org/10.1007/s12161-015-0160-4]

Wang L G and Du X P. 2017. Semi-supervised classification of hyperspectral images applying the combination of K-mean clustering and twin support vector machine. Applied Science and Technology, 44(3): 12-18

王立国, 杜心平. 2017. K均值聚类和孪生支持向量机相结合的高光谱图像半监督分类. 应用科技, 44(3): 12-18 [DOI: 10.11991/yykj.201606010http://dx.doi.org/10.11991/yykj.201606010]

Wang L G, Zhao L and Liu D F. 2018. A review on the application of SVM in hyperspectral image processing. Journal of Harbin Engineering University, 39(6): 973-983

王立国, 赵亮, 刘丹凤. 2018. SVM在高光谱图像处理中的应用综述. 哈尔滨工程大学学报, 39(6): 973-983 [DOI: 10.11990/jheu.201704074http://dx.doi.org/10.11990/jheu.201704074]

Wei L F. 2017. Research on Detection Method of Maize Variety Based on Hyperspectral Image. Shenyang: Shenyang Agricultural University: 2-5

魏利峰. 2017. 玉米种子高光谱图像品种检测方法研究. 沈阳: 沈阳农业大学: 2-5

Wu X, Zhang W Z, Lu J F, Qiu Z J and He Y. 2016. Study on visual identification of corn seeds based on hyperspectral imaging technology. Spectroscopy and Spectral Analysis, 36(2): 511-514

吴翔, 张卫正, 陆江锋, 裘正军, 何勇. 2016. 基于高光谱技术的玉米种子可视化鉴别研究. 光谱学与光谱分析, 36(2): 511-514) [DOI: 10.3964/j.issn.1000-0593(201602-0511-04]

Yang J X, Zhao Y Q and Chan J C W. 2017. Learning and transferring deep joint spectral-spatial features for hyperspectral classification. IEEE Transactions on Geoscience and Remote Sensing, 55(8): 4729-4742 [DOI: 10.1109/TGRS.2017.2698503http://dx.doi.org/10.1109/TGRS.2017.2698503]

Zhang H K, Li Y and Jiang Y N. 2018. Deep learning for hyperspectral imagery classification: the state of the art and prospects. Acta Automatica Sinica, 44(6): 961-977

张号逵, 李映, 姜晔楠. 2018. 深度学习在高光谱图像分类领域的研究现状与展望. 自动化学报, 44(6): 961-977 [DOI: 10.16383/j.aas.2018.c170190http://dx.doi.org/10.16383/j.aas.2018.c170190]

Zhao W D，Li S S，Li A，Zhang B and Chen J. 2021. Deep fusion of hyperspectral images and multi-source remote sensing data for classification with convolutional neural network. National Remote Sensing Bulletin，25（7）：1489- 1502

赵伍迪,李山山,李安,张兵,陈俊. 2021. 结合深度学习的高光谱与多源遥感数据融合分类. 遥感学报, 25(07): 1489-1502 [DOI: 10.11834/jrs.20219117http://dx.doi.org/10.11834/jrs.20219117]

Zhao Y Y, Zhu S S, Zhang C, Feng X P, Feng L and He Y. 2018. Application of hyperspectral imaging and chemometrics for variety classification of maize seeds. RSC Advances, 8(3): 1337-1345 [DOI: 10.1039/C7RA05954Jhttp://dx.doi.org/10.1039/C7RA05954J]

Alert me when the article has been cited

提交

MtSCCD： Land-use scene classification and change-detection dataset for deep learning

Joint classification of hyperspectral and LiDAR data based on inter-modality match learning

Hyperspectral unmixing network considering spectral information and superpixel segmentation

Hyperspectral remote sensing image classification based on multidirectional adaptive aware network

Spectral-spatial attention bilateral network for hyperspectral image classification