Rapid discrimination of quality grade of black tea based on near-infrared spectroscopy (NIRS), electronic nose (E-nose) and data fusion.

For the manufacturing and sale of tea, rapid discrimination of overall quality grade is of great importance. However, present evaluation methods are time-consuming and labor-intensive. This study investigated the feasibility of combining advantages of near-infrared spectroscopy (NIRS) and electronic nose (E-nose) to assess the tea quality. We found that NIRS and E-nose models effectively identify taste and aroma quality grades, with the highest accuracies of 99.63% and 97.00%, respectively, by comparing different principal component numbers and classification algorithms. Additionally, the quantitative models based on NIRS predicted the contents of key substances. Based on this, NIRS and E-nose data were fused in the feature-level to build the overall quality evaluation model, achieving accuracies of 98.13%, 96.63% and 97.75% by support vector machine, K-nearest neighbors, and artificial neural network, respectively. This study reveals that the integration of NIRS and E-nose presents a novel and effective approach for rapidly identifying tea quality.

Weed target detection at seedling stage in paddy fields based on YOLOX.

Weeds are one of the greatest threats to the growth of rice, and the loss of crops is greater in the early stage of rice growth. Traditional large-area spraying cannot selectively spray weeds and can easily cause herbicide waste and environmental pollution. To realize the transformation from large-area spraying to precision spraying in rice fields, it is necessary to quickly and efficiently detect the distribution of weeds. Benefiting from the rapid development of vision technology and deep learning, this study applies a computer vision method based on deep-learning-driven rice field weed target detection. To address the need to identify small dense targets at the rice seedling stage in paddy fields, this study propose a method for weed target detection based on YOLOX, which is composed of a CSPDarknet backbone network, a feature pyramid network (FPN) enhanced feature extraction network and a YOLO Head detector. The CSPDarknet backbone network extracts feature layers with dimensions of 80 pixels ⊆ 80 pixels, 40 pixels ⊆ 40 pixels and 20 pixels ⊆ 20 pixels. The FPN fuses the features from these three scales, and YOLO Head realizes the regression of the object classification and prediction boxes. In performance comparisons of different models, including YOLOv3, YOLOv4-tiny, YOLOv5-s, SSD and several models of the YOLOX series, namely, YOLOX-s, YOLOX-m, YOLOX-nano, and YOLOX-tiny, the results show that the YOLOX-tiny model performs best. The mAP, F1, and recall values from the YOLOX-tiny model are 0.980, 0.95, and 0.983, respectively. Meanwhile, the intermediate variable memory generated during the model calculation of YOLOX-tiny is only 259.62 MB, making it suitable for deployment in intelligent agricultural devices. However, although the YOLOX-tiny model is the best on the dataset in this paper, this is not true in general. The experimental results suggest that the method proposed in this paper can improve the model performance for the small target detection of sheltered weeds and dense weeds at the rice seedling stage in paddy fields. A weed target detection model suitable for embedded computing platforms is obtained by comparing different single-stage target detection models, thereby laying a foundation for the realization of unmanned targeted herbicide spraying performed by agricultural robots.

Modeling and De-Noising for Nondestructive Detection of Total Soluble Solid Content of Pomelo by Using Visible/Near Infrared Spectroscopy.

The flavor of Pomelo is highly variable and difficult to determine without peeling the fruit. The quality of pomelo flavor is due largely to the total soluble solid content (TSSC) in the fruit and there is a commercial need for a quick but nondestructive TSSC detection method for the industrial grading of pomelo. Due to the large size and thick mesocarp of pomelo, determining the internal quality of a pomelo fruit in a nondestructive manner is difficult, and the detection accuracy is further complicated by the noise typically generated by the common methods for the internal quality detection of other fruits. Thus, the aim of this study was to determine the optimal method to accurately detect pomelo TSSC and find a de-noising model which reduces the influence of noise on the optimal method's results. After developing a full-transmission visible/near infrared (VIS/NIR) spectroscopy sampling method, the confirming experimental results showed that the optimal pomelo TSSC detection model was Savitzky Golay + standard normal variate + competitive adaptive reweighted sampling + partial least squares regression. The R2 and RMSE of the calibration set for pomelo TSSC detection were 0.8097 and 0.8508, respectively, and the R2 and RMSE of the validation set for pomelo TSSC detection were 0.8053 and 0.8888, respectively. Both reference and dark de-noising are important for pomelo internal quality detection and should be calibrated frequently to compensate for time drift. This study found that large sensor response translation noise can be reduced with an artificial horizontal shift. Data supplementation is efficient for improving the adaption of the detection model for batch differences in pomelo samples. Using this optimized de-noising model to compensate for time drift, sensor response translation, and batch differences, the developed detection method is capable of satisfying the requirements of the industry (TSSC detection R2 was equal or larger than 0.9, RMSE was less than 1). These results indicate that full-transmission VIS/NIR spectroscopy can be exploited to realize the nondestructive detection of pomelo TSSC on an industrial scale, and that the methodologies used in this study can be immediately implemented in real-world production.

Rapid Nondestructive Detection of Water Content and Granulation in Postharvest "Shatian" Pomelo Using Visible/Near-Infrared Spectroscopy.

Visible/near-infrared (VIS/NIR) spectroscopy is a powerful tool for rapid, nondestructive fruit quality detection. This technology has been widely applied for quality detection of small, thin-peeled fruit, though less so for large, thick-peeled fruit due to a weak spectral signal resulting in a reduction of accuracy. More modeling work should be focused on solving this problem. "Shatian" pomelo is a traditional Chinese large, thick-peeled fruit, and granulation and water loss are two major internal quality factors that influence its storage quality. However, there is no efficient, nondestructive detection method for measuring these factors. Thus, the VIS/NIR spectral signal detection of 120 pomelo samples during storage was performed. Information mining (singular sample elimination, data processing, feature extraction) and modeling were performed in different ways to construct the optimal method for achieving an accurate detection. Our results showed that the water content of postharvest pomelo was optimally detected using the Savitzky-Golay method (SG) plus the multiplicative scatter correction method (MSC) for data processing, genetic algorithm (GA) for feature extraction, and partial least squares regression (PLSR) for modeling (the coefficient of determination and root mean squared error of the validation set were 0.712 and 0.0488, respectively). Granulation degree was best detected using SG for data processing and PLSR for modeling (the detection accuracy of the validation set was 100%). Additionally, our research showed a weak relationship between the pomelo water content and granulation degree, which provided a reference for the existing debates. Therefore, our results demonstrated that VIS/NIR combined with optimal information mining and modeling methodswas feasible for determining the water content and granulation degree of postharvest pomelo, and for providing references for the nondestructive internal quality detection of other large, thick-peeled fruits.

Single-Kernel FT-NIR Spectroscopy for Detecting Supersweet Corn (Zea mays L. Saccharata Sturt) Seed Viability with Multivariate Data Analysis.

The viability and vigor of crop seeds are crucial indicators for evaluating seed quality, and high-quality seeds can increase agricultural yield. The conventional methods for assessing seed viability are time consuming, destructive, and labor intensive. Therefore, a rapid and nondestructive technique for testing seed viability has great potential benefits for agriculture. In this study, single-kernel Fourier transform near-infrared (FT-NIR) spectroscopy with a wavelength range of 1000-2500 nm was used to distinguish viable and nonviable supersweet corn seeds. Various preprocessing algorithms coupled with partial least squares discriminant analysis (PLS-DA) were implemented to test the performance of classification models. The FT-NIR spectroscopy technique successfully differentiated viable seeds from seeds that were nonviable due to overheating or artificial aging. Correct classification rates for both heat-damaged kernels and artificially aged kernels reached 98.0%. The comprehensive model could also attain an accuracy of 98.7% when combining heat-damaged samples and artificially aged samples into one category. Overall, the FT-NIR technique with multivariate data analysis methods showed great potential capacity in rapidly and nondestructively detecting seed viability in supersweet corn.