The Rapid Non-Destructive Differentiation of Different Varieties of Rice by Fluorescence Hyperspectral Technology Combined with Machine Learning.

A rice classification method for the fast and non-destructive differentiation of different varieties is significant in research at present. In this study, fluorescence hyperspectral technology combined with machine learning techniques was used to distinguish five rice varieties by analyzing the fluorescence hyperspectral features of Thai jasmine rice and four rice varieties with a similar appearance to Thai jasmine rice in the wavelength range of 475-1000 nm. The fluorescence hyperspectral data were preprocessed by a first-order derivative (FD) to reduce the background and baseline drift effects of the rice samples. Then, a principal component analysis (PCA) and t-distributed stochastic neighborhood embedding (t-SNE) were used for feature reduction and 3D visualization display. A partial least squares discriminant analysis (PLS-DA), BP neural network (BP), and random forest (RF) were used to build the rice classification models. The RF classification model parameters were optimized using the gray wolf algorithm (GWO). The results show that FD-t-SNE-GWO-RF is the best model for rice classification, with accuracy values of 99.8% and 95.3% for the training and test sets, respectively. The fluorescence hyperspectral technique combined with machine learning is feasible for classifying rice varieties.

Surface Enhancement Effects of Tiny SnO2 Nanoparticle Modification on α-Fe2O3 for Room-Temperature NH3 Sensing.

The development of a gas sensor capable of detecting ammonia with high selectivity and rapid response at room temperature has consistently posed a formidable challenge. To address this issue, the present study utilized a one-step solvothermal method to co-assemble α-Fe2O3 and SnO2 by evenly covering SnO2 nanoparticles on the surface of α-Fe2O3. By controlling the morphology and Fe/Sn mole ratio of the composite, the as-prepared sample exhibits high-performance detection of NH3. At room temperature conditions, a gas sensor composed of α-Fe2O3@3%SnO2 demonstrates a rapid response time of 14 s and a notable sensitivity of 83.9% when detecting 100 ppm ammonia. Experiments and density functional theory (DFT) calculations suggest that the adsorption capacity of α-Fe2O3 to ammonia is enhanced by the surface effect provided by SnO2. Meanwhile, the existence of SnO2 tailors the pore structure and effective surface area of α-Fe2O3, creating multiple channels for the diffusion and adsorption of ammonia molecules. Additionally, an N-N heterostructure is formed between α-Fe2O3 and SnO2, which enhances the potential energy barrier and improves the ammonia sensing performance. Demonstration experiments have proved that the sensor shows significant advantages over commercial sensors in the process of ammonia detection in agricultural facilities. This work provides new insights into the perspectives on ammonia detection at room temperature.

ASFL-YOLOX: an adaptive spatial feature fusion and lightweight detection method for insect pests of the Papilionidae family.

Introduction: Insect pests from the family Papilionidae (IPPs) are a seasonal threat to citrus orchards, causing damage to young leaves, affecting canopy formation and fruiting. Existing pest detection models used by orchard plant protection equipment lack a balance between inference speed and accuracy. Methods: To address this issue, we propose an adaptive spatial feature fusion and lightweight detection model for IPPs, called ASFL-YOLOX. Our model includes several optimizations, such as the use of the Tanh-Softplus activation function, integration of the efficient channel attention mechanism, adoption of the adaptive spatial feature fusion module, and implementation of the soft Dlou non-maximum suppression algorithm. We also propose a structured pruning curation technique to eliminate unnecessary connections and network parameters. Results: Experimental results demonstrate that ASFL-YOLOX outperforms previous models in terms of inference speed and accuracy. Our model shows an increase in inference speed by 29 FPS compared to YOLOv7-x, a higher mAP of approximately 10% than YOLOv7-tiny, and a faster inference frame rate on embedded platforms compared to SSD300 and Faster R-CNN. We compressed the model parameters of ASFL-YOLOX by 88.97%, reducing the number of floating point operations per second from 141.90G to 30.87G while achieving an mAP higher than 95%. Discussion: Our model can accurately and quickly detect fruit tree pest stress in unstructured orchards and is suitable for transplantation to embedded systems. This can provide technical support for pest identification and localization systems for orchard plant protection equipment.

Design of an efficient combined multipoint picking scheme for tea buds.

Herein, a combined multipoint picking scheme was proposed, and the sizes of the end of the bud picker were selectively designed. Firstly, the end of the bud picker was abstracted as a fixed-size picking box, and it was assumed that the tea buds in the picking box have a certain probability of being picked. Then, the picking box coverage and the greedy algorithm were designed to make as few numbers of picking box set as possible to cover all buds to reduce the numbers of picking. Furthermore, the Graham algorithm and the minimum bounding box were applied to fine-tune the footholds of each picking box in the optimal coverage picking box set, so that the buds were concentrated in the middle of the picking boxes as much as possible. Moreover, the geometric center of each picking box was taken as a picking point, and the ant colony algorithm was used to optimize the picking path of the end of the bud picker. Finally, by analyzing the influence of several parameters on the picking performance of the end of the bud picker, the optimal sizes of the picking box were calculated successfully under different conditions. The experimental results showed that the average picking numbers of the combined multipoint picking scheme were reduced by 31.44%, the shortest picking path was decreased by 11.10%, and the average consumed time was reduced by 50.92% compared to the single-point picking scheme. We believe that the proposed scheme can provide key technical support for the subsequent design of intelligent bud-picking robots.

Distinguishing Different Varieties of Oolong Tea by Fluorescence Hyperspectral Technology Combined with Chemometrics.

Oolong tea is a semi-fermented tea that is popular among people. This study aims to establish a classification method for oolong tea based on fluorescence hyperspectral technology(FHSI) combined with chemometrics. First, the spectral data of Tieguanyin, Benshan, Maoxie and Huangjingui were obtained. Then, standard normal variation (SNV) and multiple scatter correction (MSC) were used for preprocessing. Principal component analysis (PCA) was used for data visualization, and with tolerance ellipses that were drawn according to Hotelling, outliers in the spectra were removed. Variable importance for the projection (VIP) > 1 in partial least squares discriminant analysis (PLS−DA) was used for feature selection. Finally, the processed spectral data was entered into the support vector machine (SVM) and PLS−DA. MSC_VIP_PLS−DA was the best model for the classification of oolong tea. The results showed that the use of FHSI could accurately distinguish these four types of oolong tea and was able to identify the key wavelengths affecting the tea classification, which were 650.11, 660.29, 665.39, 675.6, 701.17, 706.31, 742.34 and 747.5 nm. In these wavelengths, different kinds of tea have significant differences (p < 0.05). This study could provide a non-destructive and rapid method for future tea identification.

The Rapid Non-Destructive Detection of Adulteration and Its Degree of Tieguanyin by Fluorescence Hyperspectral Technology.

Tieguanyin is one of the top ten most popular teas and the representative of oolong tea in China. In this study, a rapid and non-destructive method is developed to detect adulterated tea and its degree. Benshan is used as the adulterated tea, which is about 0%, 10%, 20%, 30%, 40%, and 50% of the total weight of tea samples, mixed with Tieguanyin. Taking the fluorescence spectra from 475 to 1000 nm, we then established the 2-and 6-class discriminant models. The 2-class discriminant models had the best evaluation index when using SG-CARS-SVM, which can reach a 100.00% overall accuracy, 100.00% specificity, 100% sensitivity, and the least time was 1.2088 s, which can accurately identify pure and adulterated tea; among the 6-class discriminant models (0% (pure Tieguanyin), 10, 20, 30, 40, and 50%), with the increasing difficulty of adulteration, SNV-RF-SVM had the best evaluation index, the highest overall accuracy reached 94.27%, and the least time was 0.00698 s. In general, the results indicated that the two classification methods explored in this study can obtain the best effects. The fluorescence hyperspectral technology has a broad scope and feasibility in the non-destructive detection of adulterated tea and other fields.

Controlled fabrication of core-shell γ-Fe2O3@C-Reduced graphene oxide composites with tunable interfacial structure for highly efficient microwave absorption.

The design of a high-performance microwave absorbing material is highly dependent on the synergistic structural design of heterostructure and the appropriate material compositions. Herein, a series of composites of reduced graphene oxide (RGO) and core-shell structured γ-Fe2O3@C nanoparticles have been achieved by a hydrothermal and in-situ chemical vapor deposition (CVD) method. In particular, the structure of the carbon layer, including its graphitization and thickness, can be controlled by optimizing the CVD conditions, which is beneficial to tailor the impedance matching and dielectric loss. The rationally designed RGO/γ-Fe2O3@C composite has multiple electromagnetic dissipation mechanisms. The effective absorption ranges of an optimal sample at a filling rate of 20% can cover 100% X-band and 98% Ku-band at thicknesses of 3.0 mm and 2.2 mm, respectively. This finding suggested that the controllable fabrication of core-shell heterostructures could be viable approach to upgrade the microwave absorption performance of transition metal oxides.

Self-Adaptive Learning of Task Offloading in Mobile Edge Computing Systems.

Mobile edge computing (MEC) focuses on transferring computing resources close to the user's device, and it provides high-performance and low-delay services for mobile devices. It is an effective method to deal with computationally intensive and delay-sensitive tasks. Given the large number of underutilized computing resources for mobile devices in urban areas, leveraging these underutilized resources offers tremendous opportunities and value. Considering the spatiotemporal dynamics of user devices, the uncertainty of rich computing resources and the state of network channels in the MEC system, computing resource allocation in mobile devices with idle computing resources will affect the response time of task requesting. To solve these problems, this paper considers the case in which a mobile device can learn from a neighboring IoT device when offloading a computing request. On this basis, a novel self-adaptive learning of task offloading algorithm (SAda) is designed to minimize the average offloading delay in the MEC system. SAda adopts a distributed working mode and has a perception function to adapt to the dynamic environment in reality; it does not require frequent access to equipment information. Extensive simulations demonstrate that SAda achieves preferable latency performance and low learning error compared to the existing upper bound algorithms.

Computer Vision Based Automatic Recognition of Pointer Instruments: Data Set Optimization and Reading.

With the promotion of intelligent substations, more and more robots have been used in industrial sites. However, most of the meter reading methods are interfered with by the complex background environment, which makes it difficult to extract the meter area and pointer centerline, which is difficult to meet the actual needs of the substation. To solve the current problems of pointer meter reading for industrial use, this paper studies the automatic reading method of pointer instruments by putting forward the Faster Region-based Convolutional Network (Faster-RCNN) based object detection integrating with traditional computer vision. Firstly, the Faster-RCNN is used to detect the target instrument panel region. At the same time, the Poisson fusion method is proposed to expand the data set. The K-fold verification algorithm is used to optimize the quality of the data set, which solves the lack of quantity and low quality of the data set, and the accuracy of target detection is improved. Then, through some image processing methods, the image is preprocessed. Finally, the position of the centerline of the pointer is detected by the Hough transform, and the reading can be obtained. The evaluation of the algorithm performance shows that the method proposed in this paper is suitable for automatic reading of pointer meters in the substation environment, and provides a feasible idea for the target detection and reading of pointer meters.