Hyperspectral Image-Based Variety Classification of Waxy Maize Seeds by the t-SNE Model and Procrustes Analysis

Variety classification is an important step in seed quality testing. This study introduces t-distributed stochastic neighbourhood embedding (t-SNE), a manifold learning algorithm, into the field of hyperspectral imaging (HSI) and proposes a method for classifying seed varieties. Images of 800 maize kernels of eight varieties (100 kernels per variety, 50 kernels for each side of the seed) were imaged in the visible- near infrared (386.7–1016.7 nm) wavelength range. The images were pre-processed by Procrustes analysis (PA) to improve the classification accuracy, and then these data were reduced to low-dimensional space using t-SNE. Finally, Fisher’s discriminant analysis (FDA) was used for classification of the low-dimensional data. To compare the effect of t-SNE, principal component analysis (PCA), kernel principal component analysis (KPCA) and locally linear embedding (LLE) were used as comparative methods in this study, and the results demonstrated that the t-SNE model with PA pre-processing has obtained better classification results. The highest classification accuracy of the t-SNE model was up to 97.5%, which was much more satisfactory than the results of the other models (up to 75% for PCA, 85% for KPCA, 76.25% for LLE). The overall results indicated that the t-SNE model with PA pre-processing can be used for variety classification of waxy maize seeds and be considered as a new method for hyperspectral image analysis.

[1]  Abdul Rahman Othman,et al.  A Model Classification Technique for Linear Discriminant Analysis for Two Groups , 2012 .

[2]  Brian Brisco,et al.  Fisher Linear Discriminant Analysis of coherency matrix for wetland classification using PolSAR imagery , 2018 .

[3]  Chu Zhang,et al.  Rice Seed Cultivar Identification Using Near-Infrared Hyperspectral Imaging and Multivariate Data Analysis , 2013, Sensors.

[4]  Chu Zhang,et al.  Non-Destructive and Rapid Variety Discrimination and Visualization of Single Grape Seed Using Near-Infrared Hyperspectral Imaging Technique and Multivariate Analysis , 2018, Molecules.

[5]  Wentian Li,et al.  Application of t-SNE to Human Genetic Data , 2017, bioRxiv.

[6]  Fang Cheng,et al.  Spectral and Image Integrated Analysis of Hyperspectral Data for Waxy Corn Seed Variety Classification , 2015, Sensors.

[7]  Zhenjie Xiong,et al.  Use of Hyperspectral Imaging to Discriminate the Variety and Quality of Rice , 2015, Food Analytical Methods.

[8]  Sridhar Mahadevan,et al.  Manifold alignment using Procrustes analysis , 2008, ICML '08.

[9]  Huang Min,et al.  MODEL UPDATING OF HYPERSPECTRAL IMAGING DATA FOR VARIETY DISCRIMINATION OF MAIZE SEEDS HARVESTED IN DIFFERENT YEARS BY CLUSTERING ALGORITHM , 2016 .

[10]  Noel D.G. White,et al.  Feasibility of near-infrared hyperspectral imaging to differentiate Canadian wheat classes , 2008 .

[11]  R. Galvão,et al.  Classification of individual cotton seeds with respect to variety using near-infrared hyperspectral imaging , 2016 .

[12]  Wei Wang,et al.  Variety classification of maize kernels using near infrared (NIR) hyperspectral imaging , 2017 .

[13]  Dehai Zhu,et al.  Variety identification method of coated maize seeds based on near-infrared spectroscopy and chemometrics , 2015 .

[14]  Roberto Kawakami Harrop Galvão,et al.  A method for calibration and validation subset partitioning. , 2005, Talanta.

[15]  Kim-Kwang Raymond Choo,et al.  Improved t-SNE based manifold dimensional reduction for remote sensing data processing , 2018, Multimedia Tools and Applications.

[16]  Geoffrey E. Hinton,et al.  Visualizing Data using t-SNE , 2008 .

[17]  Da-Wen Sun,et al.  Application of Hyperspectral Imaging to Discriminate the Variety of Maize Seeds , 2015, Food Analytical Methods.

[18]  Yong He,et al.  Application of hyperspectral imaging and chemometrics for variety classification of maize seeds , 2018, RSC advances.

[19]  Joshua W. Shaevitz,et al.  Learning stable representations in a changing world with on-line t-SNE: proof of concept in the songbird , 2016 .

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

[21]  Karbhari V. Kale,et al.  A Research Review on Hyperspectral Data Processing and Analysis Algorithms , 2017, Proceedings of the National Academy of Sciences, India Section A: Physical Sciences.

[22]  Rashmi Chhabra,et al.  Microsatellite marker-based characterization of waxy maize inbreds for their utilization in hybrid breeding , 2017, 3 Biotech.

[23]  Yongzhao Zhan,et al.  Maximum Neighborhood Margin Discriminant Projection for Classification , 2014, TheScientificWorldJournal.

[24]  Lu Wang,et al.  Lychee Variety Discrimination by Hyperspectral Imaging Coupled with Multivariate Classification , 2014, Food Analytical Methods.

[25]  Fengchen Huang,et al.  Hyperspectral Imaging Target Detection Based on Improved Kernel Principal Component Analysis , 2012, Intell. Autom. Soft Comput..

[26]  Digvir S. Jayas,et al.  Hyperspectral imaging to classify and monitor quality of agricultural materials , 2015 .

[27]  D. Jayas,et al.  Identification of wheat classes using wavelet features from near infrared hyperspectral images of bulk samples. , 2009 .

[28]  Wang Pingping,et al.  Manifold alignment for dimension reduction and classification of multitemporal hyperspectral image , 2017 .

[29]  Laurens van der Maaten,et al.  Learning a Parametric Embedding by Preserving Local Structure , 2009, AISTATS.

[30]  Tao Wang,et al.  Fast Detection of Striped Stem-Borer (Chilo suppressalis Walker) Infested Rice Seedling Based on Visible/Near-Infrared Hyperspectral Imaging System , 2017, Sensors.

[31]  Julia Ling,et al.  Visualization of High Dimensional Turbulence Simulation Data using t-SNE. , 2017 .