Aflatoxin B₁ (AFB₁) is a very strong carcinogen, maize kernels are easily infected by this toxin during storage. Rapid and accurate identification of AFB₁ is of great significance to ensure food safety. In this study, a novel method for classification of AFB₁ in single maize kernels was developed. Four groups of maize kernel samples with different AFB₁ concentrations (10, 20, 50, and 100 ppb) were prepared by artificial inoculation of toxin. In addition, one group of maize kernel samples without AFB₁ were prepared as control, each group with 70 samples. The visible and short wave near-infrared (Vis-SWNIR) region (500–1000 nm) and long wave near-infrared (LWNIR) region (1000–2000 nm) hyperspectral images of all samples were obtained respectively, and the hyperspectral images in 500–2000 nm range was obtained after spectral pretreatment and fusion. Kennard-Stone algorithm was used to divide the samples into calibration set or prediction set. Competitive adaptive reweighted sampling (CARS) and successive projections algorithm (SPA) were used to roughly select the characteristic wavelengths of the calibration set samples, and 25 and 26 effective wavelengths were obtained respectively. Based on the roughly selected wavelengths, a method of fine selection of the characteristic wavelengths was proposed by using the gray-value difference of image (GDI), and a few number of characteristic wavelengths were further selected. Under the LDA classification model, 10 characteristic wavelengths were selected to test the prediction set and the independent verification samples, and the ideal result were obtained with an accuracy of 94.46% and 91.11%, respectively. This study provides a new approach for AFB₁ concentration classification of single maize kernels.

Maize is a globally major crop; however, the prevalence of mixed-aged seeds in the market complicates consumer selection and impedes the healthy development of the maize industry. This study introduces a novel method for identifying maize seeds of different storage ages. Seeds were categorized into three age groups: new seeds, one-year stored, and two-year stored, with 300 seeds per group. Hyperspectral images of all 900 samples were acquired using a visible and near-infrared (Vis-NIR) hyperspectral imaging system. To achieve optimal results with minimal spectral data, a feature wavelength selection algorithm based on Inter-Class Feature Differences (IFD) was proposed. When only using the selected three key wavelengths, combined with the linear discriminant analysis (LDA) algorithm, the discrimination accuracy among three different age groups reached 85.67%, while the discrimination accuracy between new and aged seeds achieved 95.33%. Compared to two commonly used variable selection algorithms—Successive Projections Algorithm (SPA) and Random Frog (RF), the proposed IFD method demonstrated superior performance when only a limited number of key wavelengths were used for modeling. These results indicate that the proposed algorithm offers an effective and efficient solution for maize seed age discrimination, showing great potential for practical application.

Dendrobium officinale is an important medicinal material in traditional Chinese medicine. Due to the superior quality and high market value of the Huoshan Dendrobium officinale, products from other regions are frequently sold as counterfeit Huoshan varieties. To investigate the authenticity identification of Huoshan Dendrobium officinale, this study utilized a hyperspectral imaging system covering the wavelength range of 900–1700 nm to collect hyperspectral images from Huoshan and other major production regions. After extracting spectral data of samples, spectral preprocessing was performed using Savitzky–Golay (SG) smoothing and first derivative (FD) transformation. By analyzing Principal Component Analysis (PCA) loading coefficients, up to 20 key wavelengths were selected, and a discriminant model was established based on the XGBoost algorithm. Model analysis revealed that when using the top 13 key wavelengths selected by PCA, the test accuracy for identifying Huoshan Dendrobium officinale reached 87.5%, effectively verifying its authenticity. This study contributes to promoting the healthy development of the market for traditional Chinese medicinal materials such as Dendrobium officinale.