Wheat Kernel Variety Identification Based on a Large Near-Infrared Spectral Dataset and a Novel Deep Learning-Based Feature Selection Method

Lei Zhou; Chu Zhang; Mohamed Farag Taha; Xinhua Wei; Yong He; Zhengjun Qiu; Yufei Liu

doi:10.3389/fpls.2020.575810

Wheat Kernel Variety Identification Based on a Large Near-Infrared Spectral Dataset and a Novel Deep Learning-Based Feature Selection Method

Front Plant Sci. 2020 Nov 10:11:575810. doi: 10.3389/fpls.2020.575810. eCollection 2020.

Authors

Lei Zhou^{1

2}, Chu Zhang^{1

2}, Mohamed Farag Taha^{1

2}, Xinhua Wei^{3

4}, Yong He^{1

2}, Zhengjun Qiu^{1

2}, Yufei Liu^{1

2}

Affiliations

¹ College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou, China.
² Key Laboratory of Spectroscopy Sensing, Ministry of Agriculture and Rural Affairs, Hangzhou, China.
³ Synergistic Innovation Center of Jiangsu Modern Agricultural Equipment and Technology, Zhenjiang, China.
⁴ School of Agricultural Equipment Engineering, Jiangsu University, Zhenjiang, China.

Abstract

Near-infrared (NIR) hyperspectroscopy becomes an emerging nondestructive sensing technology for inspection of crop seeds. A large spectral dataset of more than 140,000 wheat kernels in 30 varieties was prepared for classification. Feature selection is a critical segment in large spectral data analysis. A novel convolutional neural network-based feature selector (CNN-FS) was proposed to screen out deeply target-related spectral channels. A convolutional neural network with attention (CNN-ATT) framework was designed for one-dimension data classification. Popular machine learning models including support vector machine (SVM) and partial least square discrimination analysis were used as the benchmark classifiers. Features selected by conventional feature selection algorithms were considered for comparison. Results showed that the designed CNN-ATT produced a higher performance than the compared classifier. The proposed CNN-FS found a subset of features, which made a better representation of raw dataset than conventional selectors did. The CNN-ATT achieved an accuracy of 93.01% using the full spectra and keep its high precision (90.20%) by training on the 60-channel features obtained via the CNN-FS method. The proposed methods have great potential for handling the analyzing tasks on other large spectral datasets. The proposed feature selection structure can be extended to design other new model-based selectors. The combination of NIR hyperspectroscopic technology and the proposed models has great potential for automatic nondestructive classification of single wheat kernels.

Keywords: NIR hyperspectroscopy; attention mechanism; convolutional neural network; feature selection; wheat kernel classification.