Detection of seed purity of hybrid wheat using reflectance and transmittance hyperspectral imaging technology

Abstract

Chemical hybridization and genic male sterility systems are two main methods of hybrid wheat production; however, complete sterility of female wheat plants cannot be guaranteed owing to the influence of the growth stage and weather. Consequently, hybrid wheat seeds are inevitably mixed with few parent seeds, especially female seeds. Therefore, seed purity is a key factor in the popularization of hybrid wheat. However, traditional seed purity detection and variety identification methods are time-consuming, laborious, and destructive. Therefore, to establish a non-destructive classification method for hybrid and female parent seeds, three hybrid wheat varieties (Jingmai 9, Jingmai 11, and Jingmai 183) and their parent seeds were sampled. The transmittance and reflectance spectra of all seeds were collected via hyperspectral imaging technology, and a classification model was established using partial least squares-discriminant analysis (PLS-DA) combined with various preprocessing methods. The transmittance spectrum significantly improved the classification of hybrids and female parents compared to that obtained using reflectance spectrum. Specifically, using transmittance spectrum combined with a characteristic wavelength-screening algorithm, the Detrend-CARS-PLS-DA model was established, and the accuracy rates in the testing sets of Jingmai 9, Jingmai 11, and Jingmai 183 were 95.69%, 98.25%, and 97.25%, respectively. In conclusion, transmittance hyperspectral imaging combined with a machine learning algorithm can effectively distinguish female parent seeds from hybrid seeds. These results provide a reference for rapid seed purity detection in the hybrid production process. Owing to the non-destructive and rapid nature of hyperspectral imaging, the detection of hybrid wheat seed purity can be improved by online sorting in the future.

Advancements, limitations and challenges in hyperspectral imaging for comprehensive assessment of wheat quality: An up-to-date review

“… Liu et al (2020) proposed a new strategy to select 25 spectral standard deviation feature wavelength variables from the 415–995 nm range through contribution weights and categorize six close relative hybrid wheat varieties with a classification accuracy of 83.3 %. Higher accuracy was obtained from other seven research groups, who applied HIT to classify the different quantities and varieties of Chinese hybrid wheat kernels, using full 400–1000 nm range spectra or feature wavelengths, resulting in the classification accuracies ranging from 92.29 % to 99 % (details shown in Table 3 ), proving the great feasibility of HIT combined with chemometrics for the accurate classification and identification of wheat varieties ( Zhang et al, 2022 , Zhang et al, 2022 , Zhang et al, 2022 , Zhang et al, 2022 , Zhao et al, 2022 , Jin et al, 2022 , Que et al, 2023 , Jiang et al, 2023 , Zhao et al, 2023 , Lei et al, 2022 , Lv et al, 2022 ). In addition, Wu et al (2021) investigated the possibility of HIT to differentiate waxy wheat and three partial waxy wheats from wild-type wheat (five wheat lines), revealing a better overall classification accuracy of 98.51 % by a SVM model based on raw spectra of 930–2548 nm, better than that by other two models.…”

“…Catalase activity is important for judging grain aging ( Zhang et al, 2017 ). Eleven feature wavelengths were selected from 850 to 1700 nm range hyperspectral data to establish a high precision prediction model (R 2 P = 0.9664), based on a support vector machine (SVM) algorithm, to quantify and visualize the catalase activity changes in wheat kernels very well ( Zhang et al, 2022 , Zhang et al, 2022 , Zhang et al, 2022 , Zhang et al, 2022 ), which makes HIT a great potential to classify wheat based on the aging degree rapidly and non-destructively. Phosphorus is a key element related to cell wall and sugar content and is essential for wheat grain development.…”

“…Increasing spectral resolution of HIT to capture finer details in the unique spectral signatures associated with fungal contamination and mycotoxins will allow for more precise identification and characterization of specific contaminants. Coupled with advanced algorithms, including machine learning approaches ( Zhang et al, 2022 , Zhang et al, 2022 , Zhang et al, 2022 , Zhang et al, 2022 ), the accuracy and reliability of contamination detection using HIT can be improved. Enhancing the sensitivity and selectivity of HIT for detecting low levels of fungal contamination and mycotoxins is crucial for early detection and prevention of mycotoxin-related issues in food and agricultural products.…”

“…It is worth noting that HIT provides a detailed spectral signature for each pixel in an image, covering a broad NIR range of wavelengths, for wheat flour quality analysis, with different performance achieved for different range data, which is mainly due to the interaction between NIR spectra and different functional groups of biomacromolecules of wheat compositions. After being absorbed by the NIR light, the spectral points located at different locations produce different vibrations and overtones for these chemical groups ( Zhang et al, 2022 , Zhang et al, 2022 , Zhang et al, 2022 , Zhang et al, 2022 ). The main components of wheat flour include water, starch, protein, ash, and some functional substances, which are closely related to its nutritional and processing properties.…”

Hyperspectral imaging with machine learning for non-destructive classification of Astragalus membranaceus var. mongholicus, Astragalus membranaceus, and similar seeds

“…( Liu et al., 2022 ) PLS-DA is a typical classification method, which is considered as a supervision method to distinguish samples to the maximum extent. ( Nie et al., 2019 ; Zhang et al., 2022b ) MLP is a feedforward neural network. It maps a set of input vectors to a set of output vectors.…”