Identification of Defective Maize Seeds Using Hyperspectral Imaging Combined with Deep Learning

Seed quality affects crop yield and the quality of agricultural products, and traditional identification methods are time-consuming, complex, and irreversibly destructive. This study aims to establish a fast, non-destructive, and effective approach for defect detection in maize seeds based on hyperspectral imaging (HSI) technology combined with deep learning. Raw spectra collected from maize seeds (200 each healthy and worm-eaten) were pre-processed using detrending (DE) and multiple scattering correction (MSC) to highlight the spectral differences between samples. A convolutional neural network architecture (CNN-FES) based on a feature selection mechanism was proposed according to the importance of wavelength in the target classification task. The results show that the subset of 24 feature wavelengths selected by the proposed CNN-FES can capture important feature information in the spectral data more effectively than the conventional successive projections algorithm (SPA) and competitive adaptive reweighted sampling (CARS) algorithms. In addition, a convolutional neural network architecture (CNN-ATM) based on an attentional classification mechanism was designed for one-dimensional spectral data classification and compared with three commonly used machine learning methods, linear discriminant analysis (LDA), random forest (RF), and support vector machine (SVM). The results show that the classification performance of the designed CNN-ATM on the full wavelength does not differ much from the above three methods, and the classification accuracy is above 90% on both the training and test sets. Meanwhile, the accuracy, sensitivity, and specificity of CNN-ATM based on feature wavelength modeling can reach up to 97.50%, 98.28%, and 96.77% at the highest, respectively. The study shows that hyperspectral imaging-based defect detection of maize seed is feasible and effective, and the proposed method has great potential for the processing and analysis of complex hyperspectral data.

[1]  Jinbao Jiang,et al.  Moldy peanuts identification based on hyperspectral images and Point-centered convolutional neural network combined with embedded feature selection , 2022, Comput. Electron. Agric..

[2]  Yousef Abbaspour‐Gilandeh,et al.  Hyperspectral imaging coupled with multivariate analysis and artificial intelligence to the classification of maize kernels , 2022, International Agrophysics.

[3]  Li He,et al.  Hyperspectral Monitoring of Powdery Mildew Disease Severity in Wheat Based on Machine Learning , 2022, Frontiers in Plant Science.

[4]  Zengwei Zheng,et al.  Effective band selection of hyperspectral image by an attention mechanism-based convolutional network , 2022, RSC advances.

[5]  Jing Wang,et al.  Corn Seed Defect Detection Based on Watershed Algorithm and Two-Pathway Convolutional Neural Networks , 2022, Frontiers in Plant Science.

[6]  Yong He,et al.  Hyperspectral imaging technology combined with deep learning for hybrid okra seed identification , 2021, Biosystems Engineering.

[7]  Hongfeng Yu,et al.  HyperSeed: An End-to-End Method to Process Hyperspectral Images of Seeds , 2021, Sensors.

[8]  Hao Li,et al.  Discrimination of unsound wheat kernels based on deep convolutional generative adversarial network and near-infrared hyperspectral imaging technology. , 2021, Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy.

[9]  Yankun Peng,et al.  Discriminant analysis and comparison of corn seed vigor based on multiband spectrum , 2021, Computers and Electronics in Agriculture.

[10]  Lei Pang,et al.  Feasibility study on identifying seed viability of Sophora japonica with optimized deep neural network and hyperspectral imaging , 2021, Comput. Electron. Agric..

[11]  Xi Tian,et al.  Integration of textural and spectral features of Raman hyperspectral imaging for quantitative determination of a single maize kernel mildew coupled with chemometrics. , 2021, Food chemistry.

[12]  Zhenhong Rao,et al.  Identification of rice-weevil (Sitophilus oryzae L.) damaged wheat kernels using multi-angle NIR hyperspectral data , 2021 .

[13]  Keith A. Boroevich,et al.  DeepFeature: feature selection in nonimage data using convolutional neural network , 2021, Briefings Bioinform..

[14]  Shan Zeng,et al.  Rice Seed Purity Identification Technology Using Hyperspectral Image with LASSO Logistic Regression Model , 2021, Sensors.

[15]  Limiao Deng,et al.  Aflatoxin rapid detection based on hyperspectral with 1D-convolution neural network in the pixel level. , 2021, Food chemistry.

[16]  Jiangbo Li,et al.  Application of long-wave near infrared hyperspectral imaging for determination of moisture content of single maize seed. , 2021, Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy.

[17]  Paul J. Williams,et al.  Hierarchical classification pathway for white maize, defect and foreign material classification using spectral imaging , 2021 .

[18]  Laijun Sun,et al.  Hyperspectral prediction of sugarbeet seed germination based on gauss kernel SVM. , 2021, Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy.

[19]  Jiangbo Li,et al.  Maturity determination of single maize seed by using near-infrared hyperspectral imaging coupled with comparative analysis of multiple classification models , 2020 .

[20]  Yidan Bao,et al.  Application of near-infrared hyperspectral imaging for variety identification of coated maize kernels with deep learning , 2020, Infrared physics & technology.

[21]  Liu Zhang,et al.  Non-destructive identification of slightly sprouted wheat kernels using hyperspectral data on both sides of wheat kernels , 2020 .

[22]  Chu Zhang,et al.  Wheat Kernel Variety Identification Based on a Large Near-Infrared Spectral Dataset and a Novel Deep Learning-Based Feature Selection Method , 2020, Frontiers in Plant Science.

[23]  Te Ma,et al.  Rapid and non-destructive seed viability prediction using near-infrared hyperspectral imaging coupled with a deep learning approach , 2020, Comput. Electron. Agric..

[24]  Lei Yan,et al.  Hyperspectral imaging coupled with multivariate methods for seed vitality estimation and forecast for Quercus variabilis. , 2020, Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy.

[25]  Peng Li,et al.  Prediction of Sweet Corn Seed Germination Based on Hyperspectral Image Technology and Multivariate Data Regression , 2020, Sensors.

[26]  Yong Yang,et al.  Identification of Bacterial Blight Resistant Rice Seeds Using Terahertz Imaging and Hyperspectral Imaging Combined With Convolutional Neural Network , 2020, Frontiers in Plant Science.

[27]  Jinling Zhao,et al.  Diagnosis of the Severity of Fusarium Head Blight of Wheat Ears on the Basis of Image and Spectral Feature Fusion , 2020, Sensors.

[28]  Zhengtao Li,et al.  A temporal group attention approach for multitemporal multisensor crop classification , 2020 .

[29]  Liu Zhang,et al.  Hyperspectral imaging technology combined with deep forest model to identify frost-damaged rice seeds. , 2019, Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy.

[30]  Yidan Bao,et al.  Hyperspectral imaging for seed quality and safety inspection: a review , 2019, Plant Methods.

[31]  Jun Sun,et al.  Visualizing distribution of moisture content in tea leaves using optimization algorithms and NIR hyperspectral imaging , 2019, Comput. Electron. Agric..

[32]  Yong He,et al.  Automated spectral feature extraction from hyperspectral images to differentiate weedy rice and barnyard grass from a rice crop , 2019, Comput. Electron. Agric..

[33]  Étienne Belin,et al.  Recent Applications of Multispectral Imaging in Seed Phenotyping and Quality Monitoring—An Overview , 2019, Sensors.

[34]  Jiaxin Lei,et al.  Noninvasive and Nondestructive Detection of Cowpea Bruchid within Cowpea Seeds with a Hand-Held Raman Spectrometer. , 2019, Analytical chemistry.

[35]  Yuan Zhang,et al.  THz Spectroscopic Investigation of Wheat-Quality by Using Multi-Source Data Fusion , 2018, Sensors.

[36]  Baskar Ganapathysubramanian,et al.  Hyperspectral band selection using genetic algorithm and support vector machines for early identification of charcoal rot disease in soybean stems , 2018, Plant Methods.

[37]  Weijun Li,et al.  Non-destructive identification of maize haploid seeds using nonlinear analysis method based on their near-infrared spectra , 2018, Biosystems Engineering.

[38]  Wenqian Huang,et al.  Rapid and visual detection of the main chemical compositions in maize seeds based on Raman hyperspectral imaging. , 2018, Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy.

[39]  Paul J. Williams,et al.  Classification of white maize defects with multispectral imaging. , 2018, Food chemistry.

[40]  Gang Sun,et al.  Squeeze-and-Excitation Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[41]  Martin Koch,et al.  Quality Control of Sugar Beet Seeds With THz Time-Domain Spectroscopy , 2016, IEEE Transactions on Terahertz Science and Technology.

[42]  Yong He,et al.  Application of Hyperspectral Imaging and Chemometric Calibrations for Variety Discrimination of Maize Seeds , 2012, Sensors.

[43]  Hongdong Li,et al.  Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration. , 2009, Analytica chimica acta.

[44]  Madan Gopal,et al.  Least squares twin support vector machines for pattern classification , 2009, Expert Syst. Appl..

[45]  Robert P. Sheridan,et al.  Random Forest: A Classification and Regression Tool for Compound Classification and QSAR Modeling , 2003, J. Chem. Inf. Comput. Sci..

[46]  M. C. U. Araújo,et al.  The successive projections algorithm for variable selection in spectroscopic multicomponent analysis , 2001 .

[47]  Minzan Li,et al.  Chlorophyll content estimation based on cascade spectral optimizations of interval and wavelength characteristics , 2021, Comput. Electron. Agric..

[48]  Sulaymon Eshkabilov,et al.  Hyperspectral imaging techniques for rapid detection of nutrient content of hydroponically grown lettuce cultivars , 2021, Comput. Electron. Agric..

[49]  Qingyun Du,et al.  A new attention-based CNN approach for crop mapping using time series Sentinel-2 images , 2021, Comput. Electron. Agric..

[50]  Mehdi Khojastehpour,et al.  Detection of foreign materials in cocoa beans by hyperspectral imaging technology , 2021 .

[51]  Thomas Lengauer,et al.  Permutation importance: a corrected feature importance measure , 2010, Bioinform..