Nondestructive identification of barley seeds variety using near‐infrared hyperspectral imaging coupled with convolutional neural network

Abstract

Nondestructive inspection of varietal purity of seeds plays an important role in crop improvement, agricultural production, and plant breeding. In the present study, a rapid and nondestructive technique, that is, near-infrared hyperspectral imaging (NIR-HSI) was applied to discriminate the barley seeds variety. A large dataset of 35,280 seeds was collected from different locations and years incorporating 35 Indian barley varieties (29 hulled and 6 naked barley varieties). The hyperspectral reflectance images of the ventral side and dorsal side of seeds were acquired in the nearinfrared range of 900-1700 nm. Mean spectra were extracted and pretreated by six preprocessing techniques (standard normal variate (SNV), multiplicative scatter correction (MSC), Savitzky-Golay (SG) smoothing, SG first derivative, SG second derivative, and detrending). Subsequently, raw and preprocessed spectral data were fed as input to the convolutional neural network (CNN) including traditional machine learning models (partial least squares discriminant analysis (PLS-DA), K-nearest neighbors (KNN), and support vector machines (SVM)). It was observed that the end-to-end CNN model built on raw spectra overperformed the model using the preprocessing strategies. In addition, the CNN model outperformed the three traditional models with a testing set accuracy of greater than 98%. The results demonstrated that NIR-HSI coupled with end-to-end CNN could be a robust way to quickly, accurately, and nondestructively identify the variety of barley seeds. Practical ApplicationsThe commercial price and quality of barley mainly depend upon its varietal purity. Identification of barley seeds variety is an important step to select the seeds for different purposes such as food, malt, and fodder. Traditional methods for the identification of barley seeds variety are time-consuming, expensive, and destructive. Near-infrared hyperspectral imaging, as an emerging fast and nondestructive technique, looks promising for seed quality and safety evaluation. Moreover, the convolutional neural network has a better capability to accurately discriminate spectra extracted from seeds of different varieties of barley. The results in this study can provide a reference and theoretical basis to develop a real-time inspection system for fast, accurate, and nondestructive barley seeds purity testing.

Classification of Some Barley Cultivars with Deep Convolutional Neural Networks

“…This researcher suggested that the main limitation of this work is that it requires manual adjustment of sunflower seeds before they are captured on camera. In Table 5, the results obtained in this study were compared with the studies on the classification of barley cultivars in recent years (Hailu & Meshesha 2016;Dolata & Reiner 2018;Kozlowski et al 2019;Shi et al 2021;Singh et al 2021). Dolata & Reiner (2018) reported an accuracy rate of 97.24%.…”

“…According to our research with 14 barley cultivars, the fine-tuned DenseNet-169 model was performed better than other models and the state-of-the-art literature. On the other hand, Singh et al (2021) achieved the best accuracy rate of 98.38% with CNN because of near-infrared hyperspectral imaging.…”

“…This study aims to develop an effective computer vision system to classify some barley cultivars using Deep Convolutional Neural Network and help reduce the errors of traditional methods by avoiding human intervention. Few studies have been conducted on the classification of barley cultivars using artificial neural networks (Szczypińskia et al 2015;Hailu & Meshesha 2016;Dolata & Reiner 2018;Kozlowski et al 2019;Shi et al 2021;Singh et al 2021). For this aim, a novel image dataset of the seeds were created in order to classify the 14 barley cultivars.…”

Bruise detection and classification in jujube using thermal imaging and DenseNet

“…In previous studies, this classification was mostly performed using machine vision algorithms (Liu et al, 2021;Xu et al, 2021). At present, many algorithms have been used for the classification of bruises in fruits, for example, linear discriminant analysis, decision trees, Principal component analysis, Bayesian classifiers, support vector machines, and artificial neural networks (Baranowski, Mazurek, & Pastuszka-Woźniak, 2013;Boulent, Foucher, Theau, & St-Charles, 2019;ElMasry et al, 2008;Singh, Garg, & Iyengar, 2021;Wang et al, 2011;Zheng et al, 2019).…”

Automatic hierarchical model builder

“…These methods work very well even with highly collinear data, but it is commonly known that they do not perform well when there are very many classes to separate. An example of this is presented by Singh et al 3 where 35 types of barley are analyzed using near‐infrared hyperspectral imaging with the aim to discriminate varieties. The study showed that PLS‐DA decreases abruptly in performance as the number of varieties grows.…”