Spectral intelligent detection for aflatoxin B1 via contrastive learning based on Siamese network.

Aflatoxins, harmful substances found in peanuts, corn, and their derivatives, pose significant health risks. Addressing this, the presented research introduces an innovative MSGhostDNN model, merging contrastive learning with multi-scale convolutional networks for precise aflatoxin detection. The method significantly enhances feature discrimination, achieving an impressive 97.87% detection accuracy with a pre-trained model. By applying Grad-CAM, it further refines the model to identify key wavelengths, particularly 416 nm, and focuses on 40 key wavelengths for optimal performance with 97.46% accuracy. The study also incorporates a task dimensionality reduction approach for continuous learning, allowing effective ongoing aflatoxin spectrum monitoring in peanuts and corn. This approach not only boosts aflatoxin detection efficiency but also sets a precedent for rapid online detection of similar toxins, offering a promising solution to mitigate the health risks associated with aflatoxin exposure.

A comb-brushing-type green soybean pod harvesting equipment: Design and experiment.

To solve the problem of low efficiency of manual harvesting of green soybeans and lack of adaptable harvesters, in this study, a brushing-type green soybean harvester was designed. The comb-brushing type green soybean pod harvesting equipment is composed of a front-mounted separation drum, a full-width material delivery mechanism, a negative pressure cleaning system, and a stalk-pod separation system. Based on the operation requirements of the front-mounted brushing-type detachment drum, the drum parameters, parameters of comb arrangement, and structural parameters of the comb, the force analysis in detachment was performed. By taking the pod detachment rate and damage rate as the response indexes, the rotational speed of the drum, the travel speed of the device, and teeth distance as influencing factors, a three-factor five-level orthogonal rotary combination test was carried out by the software Design-Expert. By establishing mathematical regression models for various influencing factors and evaluation indicators, conducting variance analysis and significance analysis on the response indicators of each factor, the optimal parameters were obtained at a rotational speed of teeth of 397.36 rpm/min, minimum axial teeth distance of 4.8 mm and travel speed of the device of 0.5 m/s. Field test results showed that, under the optimal parameter combination, the pod detachment rate was 94%, the damage rate was 3.04%, the harvesting efficiency was greater than 0.187 hm2/h, and impurity content was less than 7.8%, all of which met the design and usage requirements. The research results can provide a reference for the design of soybean harvesters.

Identification of Defective Maize Seeds Using Hyperspectral Imaging Combined with Deep Learning.

Seed quality affects crop yield and the quality of agricultural products, and traditional identification methods are time-consuming, complex, and irreversibly destructive. This study aims to establish a fast, non-destructive, and effective approach for defect detection in maize seeds based on hyperspectral imaging (HSI) technology combined with deep learning. Raw spectra collected from maize seeds (200 each healthy and worm-eaten) were pre-processed using detrending (DE) and multiple scattering correction (MSC) to highlight the spectral differences between samples. A convolutional neural network architecture (CNN-FES) based on a feature selection mechanism was proposed according to the importance of wavelength in the target classification task. The results show that the subset of 24 feature wavelengths selected by the proposed CNN-FES can capture important feature information in the spectral data more effectively than the conventional successive projections algorithm (SPA) and competitive adaptive reweighted sampling (CARS) algorithms. In addition, a convolutional neural network architecture (CNN-ATM) based on an attentional classification mechanism was designed for one-dimensional spectral data classification and compared with three commonly used machine learning methods, linear discriminant analysis (LDA), random forest (RF), and support vector machine (SVM). The results show that the classification performance of the designed CNN-ATM on the full wavelength does not differ much from the above three methods, and the classification accuracy is above 90% on both the training and test sets. Meanwhile, the accuracy, sensitivity, and specificity of CNN-ATM based on feature wavelength modeling can reach up to 97.50%, 98.28%, and 96.77% at the highest, respectively. The study shows that hyperspectral imaging-based defect detection of maize seed is feasible and effective, and the proposed method has great potential for the processing and analysis of complex hyperspectral data.