Hyperspectral Imaging Technology and Transfer Learning Utilized in Haploid Maize Seeds Identification

It is extremely important to correctly identify the cultivars of maize seeds in the breeding process of maize. In this paper, the transfer learning as a method of deep learning is adopted to establish a model by combining with the hyperspectral imaging technology. The haploid seeds can be recognized from large amount of diploid maize ones with great accuracy through the model. First, the information of maize seeds on each wave band is collected using the hyperspectral imaging technology, and then the recognition model is built on VGG-19 network, which is pretrained by large-scale computer vision database (Image-Net). The correct identification rate of model utilizing seed spectral images containing 256 wave bands (862.5–1704.2nm) reaches 96.32%, and the correct identification rate of the model utilizing the seed spectral images containing single-band reaches 95.75%. The identification effect of the model is significantly better than existing maize cultivars identification algorithms, such as DPLS-SVM, PCA-SVM. The experimental results show that, CNN model which is pre-trained by visible light image database can be applied to the near-infrared hyperspectral imaging-based identification of maize seeds, and high accurate identification rate can be achieved. Meanwhile, when there is small amount of data samples, it can still realize high recognition by using transfer learning. The model not only meets the requirements of breeding recognition, but also greatly reduce the cost occurred in sample collection.

[1]  Albrecht E. Melchinger,et al.  Doubled Haploids in Tropical Maize: I. Effects of Inducers and Source Germplasm on in vivo Haploid Induction Rates , 2011 .

[2]  Yoshua Bengio,et al.  How transferable are features in deep neural networks? , 2014, NIPS.

[3]  Indra D. Bhatt,et al.  Characterization of Agro-diversity by Seed Storage Protein Electrophoresis: Focus on Rice Germplasm from Uttarakhand Himalaya, India , 2010 .

[4]  Birte Boelt,et al.  Classification of Viable and Non-Viable Spinach (Spinacia Oleracea L.) Seeds by Single Seed near Infrared Spectroscopy and Extended Canonical Variates Analysis , 2011 .

[5]  Kang Tu,et al.  Measurement of moisture, soluble solids, sucrose content and mechanical properties in sugar beet using portable visible and near-infrared spectroscopy , 2015 .

[6]  Ivan Laptev,et al.  Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[7]  Aaas News,et al.  Book Reviews , 1893, Buffalo Medical and Surgical Journal.

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

[9]  Colm P. O'Donnell,et al.  Hyperspectral imaging – an emerging process analytical tool for food quality and safety control , 2007 .

[10]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[11]  Pengcheng Nie,et al.  Spectral Multivariable Selection and Calibration in Visible-Shortwave Near-Infrared Spectroscopy for Non-Destructive Protein Assessment of Spirulina Microalga Powder , 2013 .

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

[13]  Rob Fergus,et al.  Visualizing and Understanding Convolutional Networks , 2013, ECCV.

[14]  S. Chalyk,et al.  The Use of Matroclinous Maize Haploids for Recurrent Selection , 2001, Russian Journal of Genetics.

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

[16]  Renfu Lu,et al.  Hyperspectral Imaging-Based Classification and Wavebands Selection for Internal Defect Detection of Pickling Cucumbers , 2013, Food and Bioprocess Technology.

[17]  Albrecht E. Melchinger,et al.  Doubled haploids in tropical maize: II. Quantitative genetic parameters for testcross performance , 2012, Euphytica.

[18]  Yan Wang,et al.  Genetic purity testing of F1 hybrid seed with molecular markers in cabbage (Brassica oleracea var. capitata) , 2013 .

[19]  Albrecht E. Melchinger,et al.  Rapid and accurate identification of in vivo-induced haploid seeds based on oil content in maize , 2013, Scientific Reports.

[20]  Chen Shao,et al.  Identification Haploid with High Oil Xenia Effect in Maize , 2003 .

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

[22]  Gustavo Carneiro,et al.  Unregistered Multiview Mammogram Analysis with Pre-trained Deep Learning Models , 2015, MICCAI.

[23]  P. Leroy,et al.  Molecular and morphological evaluation of doubled haploid lines in maize. 1. Homogeneity within DH lines , 1993, Theoretical and Applied Genetics.

[24]  J. O. Rawlings,et al.  Appropriate characters for racial classification in maize , 2008, Economic Botany.

[25]  Min Huang,et al.  Classification of maize seeds of different years based on hyperspectral imaging and model updating , 2016, Comput. Electron. Agric..