THz Spectroscopic Investigation of Wheat-Quality by Using Multi-Source Data Fusion

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.

Non-destructive Detection of Insect Foreign Bodies in Finishing Tea Product Based on Terahertz Spectrum and Image

Non-destructive testing of low-density and organic foreign bodies is the main challenge for food safety control. Terahertz time-domain spectroscopy (THz-TDS) and imaging technologies were applied to explore the feasibility of detection for insect foreign bodies in the finishing tea products. THz-TDS of tea leaves and foreign bodies of insects demonstrated significant differences in terms of time domain and frequency signals in the range of 0.3–1.0 THz. These signals were corrected by the use of adaptive iteratively reweighted penalized least squares (AirPLS), asymmetric least squares (AsLS), and baseline estimation and de-noising using sparsity (BEADS) for reducing baseline drift and enhancing effective spectral information. The K-nearest neighbor (KNN) and partial least squares discrimination analysis (PLS-DA) models showed the best performance after AirPLS correction with the prediction accuracy of 98 and 100%, respectively. In addition, the locations and outlines of insect bodies could be clearly presented via the THz-TDS image. These results suggested that THz-TDS spectroscopy and imaging provide an alternative tool for the detection of insect foreign bodies in finishing tea products.

The Application of Terahertz Time-Domain Spectroscopy to Identification of Potato Late Blight and Fusariosis

Fusarium and late blight (fungal diseases of cereals and potatoes) are among the main causes of crop loss worldwide. A key element of success in the fight against phytopathogens is the timely identification of infected plants and seeds. That is why the development of new methods for identifying phytopathogens is a priority for agriculture. The terahertz time-domain spectroscopy (THz-TDS) is a promising method for assessing the quality of materials. For the first time, we used THz-TDS for assessing the infection of seeds of cereals (oats, wheat and barley) with fusarium and potato tubers of different varieties (Nadezhda and Meteor) with late blight. We evaluated the refractive index, absorption coefficient and complex dielectric permittivity in healthy and infected plants. The presence of phytopathogens on seeds was confirmed by microscopy and PCR. It is shown, that Late blight significantly affected all the studied spectral characteristics. The nature of the changes depended on the variety of the analyzed plants and the localization of the analyzed tissue relative to the focus of infection. Fusarium also significantly affected all the studied spectral characteristics. It was found that THz-TDS method allows you to clearly establish the presence or absence of a phytopathogens, in the case of late blight, to assess the degree and depth of damage to plant tissues.

Determination of wheat moisture using terahertz spectroscopy combined with the tabu search algorithm

The detection of the wheat moisture content plays a key role in grain storage and classification. Harvested wheat grains were taken as samples in the current research. A total of 240 reaped wheat samples with different moisture contents were tested by applying terahertz (THz) spectroscopy. The frequency domain spectra and absorption coefficient spectra of wheat were obtained in the band of 0.1-1.2 THz, and the spectra were pretreated by mean centering, Savitzky-Golay (S-G), Multiplicative Scatter Correction (MSC) and Stand Normal Variate (SNV), respectively. Then a special algorithm of Tabu Search (TS) was used to find out the effective variables and remove the useless variables from the terahertz spectrum of the sample. Finally, the partial least squares (PLS) of chemometrics were used for quantitative model building and prediction. The correlation coefficient of calibration (R_c) is 0.9522. The root mean square error of calibration (RMSEC) is 0.4730. The correlation coefficient of prediction (R_p) is 0.9531. The root mean square error of prediction (RMSEP) is 0.5396. The results demonstrated that an accurate quantitative analysis of moisture in wheat samples could be achieved by terahertz time-domain spectroscopy combined with the TS algorithm. In addition, the results show that the model S-G + MSC + TS + PLS can effectively predict wheat moisture, and provide a rapid quantitative detection and analysis method for the detection of wheat moisture.

Assessing frost damage in barley using terahertz imaging

Frost is estimated to cost Australian grain growers $ 360 million in direct and indirect losses every year. Assessing frost damage manually in barley is labor intensive and involves destructive sampling. To mitigate against significant economic loss, it is crucial that assessment decisions on whether to cut for hay or continue to harvest are made soon after frost damage has occurred. In this paper, we propose a non-destructive technique by using raster-scan terahertz imaging. Terahertz waves can penetrate the spike to determine differences between frosted and unfrosted grains. With terahertz raster-scan imaging, conducted in both transmission and reflection at 275 GHz, frosted and unfrosted barley spikes show significant differences. In addition, terahertz imaging allows to determine individual grain positions. The emergence of compact terahertz sources and cameras would enable field deployment of terahertz non-destructive inspection for early frost damage.