Discrimination of internal crack for rice seeds using near infrared spectroscopy.

It is an important thing to identify internal crack in seeds from normal seeds for evaluating the quality of rice seeds (Oryza sativa L.). In this study, non-destructive discrimination of internal crack in rice seeds using near infrared spectroscopy and chemometrics is proposed. Principal component analysis (PCA) was used to analyze the rice seeds spectra. Four supervised classification techniques(partial least squares discriminate analysis (PLS-DA), support vector machines (SVM), k-nearest neighbors (KNN) and random forest (RF)) with four different pre-processing techniques (standard normal variate (SNV), multiplicative scatter correction (MSC), first and second derivative with Savitzky-Golay (SG) smoothing) were applied. The best results (Sn = 0.8824, Sp = 0.9429, Acc = 0.913) were achieved by PLS-DA with the raw spectral data. The performance of the best SVM model was inferior to that of PLS-DA, but superior to that of RF and KNN. Except for PLS-DA, four different preprocessing techniques were improved the performance of the developed models. The important variables for discriminating internal cracks in rice seeds were related to the amylose. Overall, the all results demonstrated the feasibility of non-destructive discrimination of internal crack for rice seeds (Oryza sativa L.) using near infrared spectroscopy and chemometrics.

Underwater image enhancement using Divide-and-Conquer network.

Underwater image enhancement has become the requirement for more people to have a better visual experience or to extract information. However, underwater images often suffer from the mixture of color distortion and blurred quality degradation due to the external environment (light attenuation, background noise and the type of water). To solve the above problem, we design a Divide-and-Conquer network (DC-net) for enhancing underwater image, which mainly consists of a texture network, a color network and a refinement network. Specifically, the multi-axis attention block is presented in the texture network, which combine different region/channel features into a single stream structure. And the color network employs an adaptive 3D look-up table method to obtain the color enhanced results. Meanwhile, the refinement network is presented to focus on image features of ground truth. Compared to state-of-the-art (SOTA) underwater image enhance methods, our proposed method can obtain the better visual quality of underwater images and better qualitative and quantitative performance. The code is publicly available at https://github.com/zhengshijian1993/DC-Net.

A wheat spike detection method based on Transformer.

Wheat spike detection has important research significance for production estimation and crop field management. With the development of deep learning-based algorithms, researchers tend to solve the detection task by convolutional neural networks (CNNs). However, traditional CNNs equip with the inductive bias of locality and scale-invariance, which makes it hard to extract global and long-range dependency. In this paper, we propose a Transformer-based network named Multi-Window Swin Transformer (MW-Swin Transformer). Technically, MW-Swin Transformer introduces the ability of feature pyramid network to extract multi-scale features and inherits the characteristic of Swin Transformer that performs self-attention mechanism by window strategy. Moreover, bounding box regression is a crucial step in detection. We propose a Wheat Intersection over Union loss by incorporating the Euclidean distance, area overlapping, and aspect ratio, thereby leading to better detection accuracy. We merge the proposed network and regression loss into a popular detection architecture, fully convolutional one-stage object detection, and name the unified model WheatFormer. Finally, we construct a wheat spike detection dataset (WSD-2022) to evaluate the performance of the proposed methods. The experimental results show that the proposed network outperforms those state-of-the-art algorithms with 0.459 mAP (mean average precision) and 0.918 AP50. It has been proved that our Transformer-based method is effective to handle wheat spike detection under complex field conditions.

Determination of soil pH from Vis-NIR spectroscopy by extreme learning machine and variable selection: A case study in lime concretion black soil.

Variable selection is widely accepted as an important step in the quantitative analysis of visible and near-infrared (Vis-NIR) spectroscopy, as it tends to improve the model's robustness and predictive ability. In this study, a total of 140 lime concretion black soil samples were collected from two towns in Guoyang County, China. The Vis-NIR spectra measured in the laboratory were used to estimate soil pH by an extreme learning machine (ELM). First, the soil spectra were treated by the optimized continuous wavelet transform (CWT), and then four spectral feature selection methods (competitive adaptive reweighted sampling, CARS; successive projections algorithm, SPA; Monte Carlo uninformative variable elimination, MCUVE; genetic algorithm, GA) were applied with ELM in the CWT domain to determine the techniques with most predictions. For comparison, The PLS and SVM models were also developed. The coefficient of determination (R2), root mean square error (RMSE), and residual prediction deviation (RPD) were used to evaluate the model performance. Based on the validation dataset, the performance of the ELM models was superior to that of the PLS and SVM models expect SPA and MCUVE. In the ELM models, the order of the prediction accuracy was GA-ELM (R2p = 0.86; RMSEp = 0.1484; RPD = 2.64), CARS-ELM (R2p = 0.84; RMSEp = 0.1565; RPD = 2.50), ELM (R2p = 0.84; RMSEp = 0.1572; RPD = 2.49), SPA-ELM (R2p = 0.84; RMSEp = 0.1589; RPD = 2.47) and MCUVE-ELM (R2p = 0.83; RMSEp = 0.1599; RPD = 2.45). The proposed method of CARS-ELM had a relatively strong ability for spectral variable selection while retaining excellent prediction accuracy and short computing time (0.39 s). In addition, the variables selected by the four methods (CARS, SPA, MCUVE and GA) indicated the prediction mechanism for pH in lime concretion black soil may be the relation between pH and iron oxides and organic matter. In conclusion, CARS-ELM has great potential to accurately determine the pH in lime concretion black soil using Vis-NIR spectroscopy.

Fast location and segmentation of high-throughput damaged soybean seeds with invertible neural networks.

BACKGROUND: Fast identification of damaged soybean seeds has undeniable importance in seed sorting and food quality. Mechanical vibration is generally used in soybean seed sorting, but this can seriously damage soybean seeds. The convolutional neural network (CNN) is considered an effective method for location and segmentation tasks. However, a CNN requires a large amount of ground truth data and has high computational cost. RESULTS: First, we propose a self-supervision manner to automatically generate ground truths, which can theoretically create an almost unlimited number of labeled images. Second, instead of using popular CNNs, a novel invertible convolution (involution)-enabled scheme is proposed by using the bottleneck block of the residual networks. Third, a feature selection feature pyramid network (FS-FPN) based on involution is designed, which selects features more flexibly and adaptively. We further merge involution-based backbones and FS-FPN into a unified network, achieving an end-to-end seed location and segmentation model; the best mean average precision of location and segmentation achieved was 85.1% and 81% respectively. CONCLUSION: The experimental results demonstrate that the proposed method greatly improves the performance of the baseline network with faster speed and fewer parameters, enabling it to detect soybean seeds more effectively. © 2022 Society of Chemical Industry.

Reflectance Spectroscopy with Multivariate Methods for Non-Destructive Discrimination of Edible Oil Adulteration.

The visible and near-infrared (Vis-NIR) reflectance spectroscopy was utilized for the rapid and nondestructive discrimination of edible oil adulteration. In total, 110 samples of sesame oil and rapeseed oil adulterated with soybean oil in different levels were produced to obtain the reflectance spectra of 350-2500 nm. A set of multivariant methods was applied to identify adulteration types and adulteration rates. In the qualitative analysis of adulteration type, the support vector machine (SVM) method yielded high overall accuracy with multiple spectra pretreatments. In the quantitative analysis of adulteration rate, the random forest (RF) combined with multivariate scattering correction (MSC) achieved the highest identification accuracy of adulteration rate with the full wavelengths of Vis-NIR spectra. The effective wavelengths of the Vis-NIR spectra were screened to improve the robustness of the multivariant methods. The analysis results suggested that the competitive adaptive reweighted sampling (CARS) was helpful for removing the redundant information from the spectral data and improving the prediction accuracy. The PLSR + MSC + CARS model achieved the best prediction performance in the two adulteration cases of sesame oil and rapeseed oil. The coefficient of determination (RPcv2) and the root mean square error (RMSEPcv) of the prediction set were 0.99656 and 0.01832 in sesame oil adulterated with soybean oil, and the RPcv2 and RMSEPcv were 0.99675 and 0.01685 in rapeseed oil adulterated with soybean oil, respectively. The Vis-NIR reflectance spectroscopy with the assistance of multivariant analysis can effectively discriminate the different adulteration rates of edible oils.

Rapid Detection of Fatty Acids in Edible Oils Using Vis-NIR Reflectance Spectroscopy with Multivariate Methods.

The composition and content of fatty acids are critical indicators to identify the quality of edible oils. This study was undertaken to establish a rapid determination method for quality detection of edible oils based on quantitative analysis of palmitic acid, stearic acid, arachidic acid, and behenic acid. Seven kinds of oils were measured to obtain Vis-NIR spectra. Multivariate methods combined with pretreatment methods were adopted to establish quantitative analysis models for the four fatty acids. The model of support vector machine (SVM) with standard normal variate (SNV) pretreatment showed the best predictive performance for the four fatty acids. For the palmitic acid, the determination coefficient of prediction (RP2) was 0.9504 and the root mean square error of prediction (RMSEP) was 0.8181. For the stearic acid, RP2 and RMSEP were 0.9636 and 0.2965. In the prediction of arachidic acid, RP2 and RMSEP were 0.9576 and 0.0577. In the prediction of behenic acid, the RP2 and RMSEP were 0.9521 and 0.1486. Furthermore, the effective wavelengths selected by successive projections algorithm (SPA) were useful for establishing simplified prediction models. The results demonstrate that Vis-NIR spectroscopy combined with multivariate methods can provide a rapid and accurate approach for fatty acids detection of edible oils.

Univariate and multivariate analyses of strontium and vanadium in soil by laser-induced breakdown spectroscopy.

Univariate and multivariate analyses of strontium (Sr) and vanadium (V) elements in soil have been performed using laser-induced breakdown spectroscopy technology. Thirty-three samples were used as a calibration set, and 11 samples were used as a prediction set. The results demonstrated that the correlation coefficients of the calibration curves method were poor due to the matrix effect. Then, the multivariate models of partial least-squares regression and least squares support vector regression (LS-SVR) were used to construct models. The analysis accuracy was improved effectively by the LS-SVR method, and the correlation coefficient is 0.999 for Sr and 0.983 for V. The average relative errors for the prediction set are lower than 7.45% and 2.88% for Sr and V, respectively. The results indicated that the LIBS technique coupled with LS-SVR could be a reliable and accurate method in the quantitative determination of elemental Sr and V in complex matrices like soil.

Accurate and rapid detection of soil and fertilizer properties based on visible/near-infrared spectroscopy.

Accurate information of soil macronutrient contents and fertilizer macronutrient contents is the precondition of precision fertilization; however, how to detect soil and fertilizer information rapidly, reliably, and inexpensively remains a great challenge. Visible and near-infrared (VIS/NIR) diffuse reflectance spectroscopy proves to be an effective tool for extensive investigation of soil and fertilizer properties. This study first collected many soil and chemical fertilizer samples and performed both spectral scanning and chemical analysis. During the correlation between the collected VIS/NIR spectra and the measured data, different spectral pretreatment, sample selection, and wavelength optimization methods were applied for improving the accuracy and robustness of the prediction models. After appropriate spectral processing and selection of representative samples, both principal component regression and genetic algorithm (GA) can adequately reduce the number of variables and pick out the characteristic variables, which not only enhanced prediction speed but also greatly improved prediction accuracy. In particular, using GA-based models, organic matter content (OMC), total N and pH value in soil and N, P, and K contents in fertilizer can all be accurately predicted.

[Research on high resolution spectrum measurements of methane at 1.65 microm].

The high resolution spectrum of methane was obtained around 1.65 microm using a tunable DFB diode laser with a long adjustable optical path white * cell (46.36-1 158.90 m) at room temperature through the direct absorption technique. The typical line width of the DFB diode laser is about 10 MHz and the wavelength of DFB laser was calibrated by an optical wavemeter. A total of 259 new absorption lines were studied from 6 043.00 to 6 053.72 cm(-1) at five different pressures and optical lengths. All the data were fitted by Gaussian profile, the line intensities, positions and the percent of the statistical standard deviation (sigmaS/ S) % of the line intensities were obtained, and the absorption lines which are hard to be distinguished were analyzed in this paper. The weakest absorption line is 4.3 x 10(-27) cm(-1) x (mol x cm(-2))(-1), while the lines stronger than 3.0 x 10(-24) cm(-1) x (mol x cm(-2))(-1) were ignored for their saturated absorption due to the long absorption optical length(788-1 000 m). Meantime, the spectrum shows the abundance of methane weak lines and its extremely complex structure around 1.65 microm. All the lines cannot be found in HITRAN2004 database, and as to our knowledge, they were not reported before by other papers.