Single Seed Identification in Three Medicago Species via Multispectral Imaging Combined with Stacking Ensemble Learning

Abstract

Multispectral imaging (MSI) has become a new fast and non-destructive detection method in seed identification. Previous research has usually focused on single models in MSI data analysis, which always employed all features and increased the risk to efficiency and that of system cost. In this study, we developed a stacking ensemble learning (SEL) model for successfully identifying a single seed of sickle alfalfa (Medicago falcata), hybrid alfalfa (M. varia), and alfalfa (M. sativa). SEL adopted a three-layer structure, i.e., level 0 with principal component analysis (PCA), linear discriminant analysis (LDA), and quadratic discriminant analysis (QDA) as models of dimensionality reduction and feature extraction (DRFE); level 1 with support vector machine (SVM), multiple logistic regression (MLR), generalized linear models with elastic net regularization (GLMNET), and eXtreme Gradient Boosting (XGBoost) as basic learners; and level 3 with XGBoost as meta-learner. We confirmed that the values of overall accuracy, kappa, precision, sensitivity, specificity, and sensitivity in the SEL model were all significantly higher than those in basic models alone, based on both spectral features and a combination of morphological and spectral features. Furthermore, we also developed a feature filtering process and successfully selected 5 optimal features out of 33 ones, which corresponded to the contents of chlorophyll, anthocyanin, fat, and moisture in seeds. Our SEL model in MSI data analysis provided a new way for seed identification, and the feature filter process potentially could be used widely for development of a low-cost and narrow-channel sensor.

Cultivars identification of oat (Avena sativa L.) seed via multispectral imaging analysis

“…Multispectral imaging, as an emerging technology, combines imaging and spectroscopy techniques, which can not only provide spatial information of seeds such as area, perimeter, length, shape, color, etc., but also provide detailed information about chemical composition, structure and other internal characteristics, determining various traits of seeds simultaneously. The merits of non-destructive, easy and no sample pre-treatment make it widely applied to seed quality testing, such as identifying rice seeds ( Liu et al., 2016b ), classifying different tomato seed cultivars ( Shrestha et al., 2016 ), discriminating alfalfa cultivars seed ( Yang et al., 2020 ; Jia et al., 2022 ), classifying Jatropha curcas seed health ( Barboza da Silva et al., 2021 ). Sumathi and Balamurugan applied machine vision technology to identify the morphology of 11 oat cultivars, and the observation was more accurate and objective than manual method ( Sumathi and Balamurugan, 2013 ).…”

Integrating optical imaging techniques for a novel approach to evaluate Siberian wild rye seed maturity

“…The main principle of the technique was based on the detection of different specific wavelengths produced by the varying physical structures and chemical compositions of objects. For example, multispectral imaging has been successfully used to identify variety genuineness and seeds quality, such as alfalfa ( Medicago sativa L.) seeds ( Yang et al., 2020 ; Jia et al., 2022 ), manioca ( Jatropha curcas L.) seeds ( Pinheiro et al., 2020 ), and spinach ( Spinacia oleracea L.) seeds ( Deleuran et al., 2013 ). The imaging technology allowed a better understanding of the seed maturation process and provided a research basis for the development of rapid, non-destructive, and high-throughput detection methods.…”