Soil data augmentation and model construction based on spectral difference and content difference.

Soil analysis makes for developing precision agriculture and monitoring land quality, while the models available for spectroscopy-based chemometrics are constrained by limited samples from small areas. The paper proposed sample expansion and model construction based on spectral difference and content difference, realizing data augmentation and deep learning applied to original samples with limited numbers. The spectral subtraction based on maximum or minimum values exploited the maximum or minimum values to acquire the spectral difference and content difference, which provided a new data form for model construction. Keeping enhanced samples whose spectral difference and content difference were all zero was useful for improving model performance. Augmentation of all data or training data based on maximum or minimum values-based spectral subtraction, which sorted the contents and made them the maximum or minimum values in sequence, achieved sample expansion by the spectral difference and content difference. The model utilized the random vector functional link (RVFL) network, extreme learning machine (ELM), and one-dimensional convolutional neural network (1D CNN), which could predict the content of new samples through ensemble averaging when predicting content difference. The experimental result showed the model of the spectral subtraction based on maximum or minimum values had a similar performance to that of the original samples. Augmentation of all data improved model performance by only RVFL and ELM. Augmentation of training data verified 1D CNN was better than RVFL and ELM. The paper implements a new data augmentation method and applies CNN to original samples with inadequate numbers, which lays the foundation for an improved model and applying spectral preprocessing.

Research on steel surface defect classification method based on deep learning.

Surface defects on steel, arising from factors like steel composition and manufacturing techniques, pose significant challenges to industrial production. Efficient and precise detection of these defects is crucial for enhancing production efficiency and product quality. In accordance with these requisites, this paper elects to undertake the detection task predicated on the you only look once (YOLO) algorithm. In this study, we propose a novel approach for surface flaw identification based on the YOLOv5 algorithm, called YOLOv5-KBS. This method integrates attention mechanism and weighted Bidirectional Feature Pyramid Network (BiFPN) into YOLOv5 architecture. Our method addresses issues of background interference and defect size variability in images. Experimental results show that the YOLOv5-KBS model achieves a notable 4.2% increase in mean Average Precision (mAP) and reaches a detection speed of 70 Frames Per Second (FPS), outperforming the baseline model. These findings underscore the effectiveness and potential applications of our proposed method in industrial settings.

An Efficient Convolutional Denoising Autoencoder-Based BDS NLOS Detection Method in Urban Forest Environments.

The BeiDou Navigation Satellite System (BDS) provides real-time absolute location services to users around the world and plays a key role in the rapidly evolving field of autonomous driving. In complex urban environments, the positioning accuracy of BDS often suffers from large deviations due to non-line-of-sight (NLOS) signals. Deep learning (DL) methods have shown strong capabilities in detecting complex and variable NLOS signals. However, these methods still suffer from the following limitations. On the one hand, supervised learning methods require labeled samples for learning, which inevitably encounters the bottleneck of difficulty in constructing databases with a large number of labels. On the other hand, the collected data tend to have varying degrees of noise, leading to low accuracy and poor generalization performance of the detection model, especially when the environment around the receiver changes. In this article, we propose a novel deep neural architecture named convolutional denoising autoencoder network (CDAENet) to detect NLOS in urban forest environments. Specifically, we first design a denoising autoencoder based on unsupervised DL to reduce the long time series signal dimension and extract the deep features of the data. Meanwhile, denoising autoencoders improve the model's robustness in identifying noisy data by introducing a certain amount of noise into the input data. Then, an MLP algorithm is used to identify the non-linearity of the BDS signal. Finally, the performance of the proposed CDAENet model is validated on a real urban forest dataset. The experimental results show that the satellite detection accuracy of our proposed algorithm is more than 95%, which is about an 8% improvement over existing machine-learning-based methods and about 3% improvement over deep-learning-based approaches.

Dynamic sparse coding-based value estimation network for deep reinforcement learning.

Deep Reinforcement Learning (DRL) is one powerful tool for varied control automation problems. Performances of DRL highly depend on the accuracy of value estimation for states from environments. However, the Value Estimation Network (VEN) in DRL can be easily influenced by the phenomenon of catastrophic interference from environments and training. In this paper, we propose a Dynamic Sparse Coding-based (DSC) VEN model to obtain precise sparse representations for accurate value prediction and sparse parameters for efficient training, which is not only applicable in Q-learning structured discrete-action DRL but also in actor-critic structured continuous-action DRL. In detail, to alleviate interference in VEN, we propose to employ DSC to learn sparse representations for accurate value estimation with dynamic gradients beyond the conventional ℓ1 norm that provides same-value gradients. To avoid influences from redundant parameters, we employ DSC to prune weights with dynamic thresholds more efficiently than static thresholds like ℓ1 norm. Experiments demonstrate that the proposed algorithms with dynamic sparse coding can obtain higher control performances than existing benchmark DRL algorithms in both discrete-action and continuous-action environments, e.g., over 25% increase in Puddle World and about 10% increase in Hopper. Moreover, the proposed algorithm can reach convergence efficiently with fewer episodes in different environments.

Rapid detection of molybdenum ore grade based on visible-infrared spectroscopy and MTSVD-TGJO-ELM.

The rapid determination of ore grade can improve the efficiency of beneficiation. The existing molybdenum ore grade determination methods lag behind the beneficiation work. Therefore, this paper proposes a method based on a combination of Visible-infrared spectroscopy and machine learning to rapidly determine molybdenum ore grade. Firstly, 128 molybdenum ores were collected as spectral test samples to obtain spectral data. Then 13 latent variables were extracted from the 973 spectral features using partial least square. The Durbin-Watson test and the runs test were used to detect the partial residual plots and augmented partial residual plots of LV1 and LV2 to determine the non-linear relationship between spectral signal and molybdenum content. Extreme Learning Machine (ELM) was used instead of linear modeling methods to model the grade of molybdenum ores because of the non-linear behavior of the spectral data. In this paper, the Golden Jackal Optimization of adaptive T-distribution was used to optimize the parameters of the ELM to solve the problem of unreasonable parameters. Aiming at solving ill-posed problems by ELM, this paper decomposes the ELM output matrix by using the improved truncated singular value decomposition. Finally, this paper proposes an extreme learning machine method based on a modified truncated singular value decomposition and a Golden Jackal Optimization of adaptive T-distribution (MTSVD-TGJO-ELM). Compared with other classical machine learning algorithms, MTSVD-TGJO-ELM has the highest accuracy. This provides a new method for rapid detection of ore grade in the mining process and facilitates accurate beneficiation of molybdenum ores to improve ore recovery rate.

Spot Detection for Laser Sensors Based on Annular Convolution Filtering.

Spot detection has attracted continuous attention for laser sensors with applications in communication, measurement, etc. The existing methods often directly perform binarization processing on the original spot image. They suffer from the interference of the background light. To reduce this kind of interference, we propose a novel method called annular convolution filtering (ACF). In our method, the region of interest (ROI) in the spot image is first searched by using the statistical properties of pixels. Then, the annular convolution strip is constructed based on the energy attenuation property of the laser and the convolution operation is performed in the ROI of the spot image. Finally, a feature similarity index is designed to estimate the parameters of the laser spot. Experiments on three datasets with different kinds of background light show the advantages of our ACF method, with comparison to the theoretical method based on international standard, the practical method used in the market products, and the recent benchmark methods AAMED and ALS.

Insights into membrane association of the SMP domain of extended synaptotagmin.

The Synaptotagmin-like Mitochondrial-lipid-binding Protein (SMP) domain is a newly identified lipid transfer module present in proteins that regulate lipid homeostasis at membrane contact sites (MCSs). However, how the SMP domain associates with the membrane to extract and unload lipids is unclear. Here, we performed in vitro DNA brick-assisted lipid transfer assays and in silico molecular dynamics simulations to investigate the molecular basis of the membrane association by the SMP domain of extended synaptotagmin (E-Syt), which tethers the tubular endoplasmic reticulum (ER) to the plasma membrane (PM). We demonstrate that the SMP domain uses its tip region to recognize the extremely curved subdomain of tubular ER and the acidic-lipid-enriched PM for highly efficient lipid transfer. Supporting these findings, disruption of these mechanisms results in a defect in autophagosome biogenesis contributed by E-Syt. Our results suggest a model that provides a coherent picture of the action of the SMP domain at MCSs.

Rapid proximate analysis of coal based on reflectance spectroscopy and deep learning.

Proximate analysis of coal is of profound significance for understanding coal quality and promoting rational utilization of coal resources. Traditional coal proximate analysis mainly uses chemical analysis methods, which have the disadvantages of slow speed and high cost. This paper proposed an approach combining reflectance spectroscopy with deep learning (DL) for rapid proximate analysis of coal. First, 80 sets of coal spectral data are enhanced by data augmentation, outlier detection, and dimensional transformation to improve the number and quality of samples. Then, an analytical model combining dilated convolution, multi-level residual connection, and a two-hidden-layer extreme learning machine (TELM), named DR_TELM, was proposed. The model extracted effective features from coal spectral data by a convolutional neural network (CNN) and utilized TELM as a regressor to achieve feature identification and content prediction. The experimental results showed that DR_TELM achieved coefficients of determination (R2) of 0.981, 0.989, 0.990, 0.985, 0.989 and root mean square errors (RMSE) of 0.533, 1.833, 1.111, 1.808, 0.723 for the content prediction of moisture, ash, volatile matter, fixed carbon and higher heating value (HHV), respectively. And while ensuring high accuracy, the test time is only 0.034 s. It is fully demonstrated that DR_TELM can rapidly and accurately analyze coal.

Direct-Optimization-Based DC Dictionary Learning With the MCP Regularizer.

Direct-optimization-based dictionary learning has attracted increasing attention for improving computational efficiency. However, the existing direct optimization scheme can only be applied to limited dictionary learning problems, and it remains an open problem to prove that the whole sequence obtained by the algorithm converges to a critical point of the objective function. In this article, we propose a novel direct-optimization-based dictionary learning algorithm using the minimax concave penalty (MCP) as a sparsity regularizer that can enforce strong sparsity and obtain accurate estimation. For solving the corresponding optimization problem, we first decompose the nonconvex MCP into two convex components. Then, we employ the difference of the convex functions algorithm and the nonconvex proximal-splitting algorithm to process the resulting subproblems. Thus, the direct optimization approach can be extended to a broader class of dictionary learning problems, even if the sparsity regularizer is nonconvex. In addition, the convergence guarantee for the proposed algorithm can be theoretically proven. Our numerical simulations demonstrate that the proposed algorithm has good convergence performances in different cases and robust dictionary-recovery capabilities. When applied to sparse approximations, the proposed approach can obtain sparser and less error estimation than the different sparsity regularizers in existing methods. In addition, the proposed algorithm has robustness in image denoising and key-frame extraction.

Accelerated Partially Shared Dictionary Learning With Differentiable Scale-Invariant Sparsity for Multi-View Clustering.

Multiview dictionary learning (DL) is attracting attention in multiview clustering due to the efficient feature learning ability. However, most existing multiview DL algorithms are facing problems in fully utilizing consistent and complementary information simultaneously in the multiview data and learning the most precise representation for multiview clustering because of gaps between views. This article proposes an efficient multiview DL algorithm for multiview clustering, which uses the partially shared DL model with a flexible ratio of shared sparse coefficients to excavate both consistency and complementarity in the multiview data. In particular, a differentiable scale-invariant function is used as the sparsity regularizer, which considers the absolute sparsity of coefficients as the l0 norm regularizer but is continuous and differentiable almost everywhere. The corresponding optimization problem is solved by the proximal splitting method with extrapolation technology; moreover, the proximal operator of the differentiable scale-invariant regularizer can be derived. The synthetic experiment results demonstrate that the proposed algorithm can recover the synthetic dictionary well with reasonable convergence time costs. Multiview clustering experiments include six real-world multiview datasets, and the performances show that the proposed algorithm is not sensitive to the regularizer parameter as the other algorithms. Furthermore, an appropriate coefficient sharing ratio can help to exploit consistent information while keeping complementary information from multiview data and thus enhance performances in multiview clustering. In addition, the convergence performances show that the proposed algorithm can obtain the best performances in multiview clustering among compared algorithms and can converge faster than compared multiview algorithms mostly.

Double Sparse Deep Reinforcement Learning via Multilayer Sparse Coding and Nonconvex Regularized Pruning.

Deep reinforcement learning (DRL), which highly depends on the data representation, has shown its potential in many practical decision-making problems. However, the process of acquiring representations in DRL is easily affected by interference from models, and moreover leaves unnecessary parameters, leading to control performance reduction. In this article, we propose a double sparse DRL via multilayer sparse coding and nonconvex regularized pruning. To alleviate interference in DRL, we propose a multilayer sparse-coding-structural network to obtain deep sparse representation for control in reinforcement learning. Furthermore, we employ a nonconvex log regularizer to promote strong sparsity, efficiently removing the unnecessary weights with a regularizer-based pruning scheme. Hence, a double sparse DRL algorithm is developed, which can not only learn deep sparse representation to reduce the interference but also remove redundant weights while keeping the robust performance. The experimental results in five benchmark environments of the deep q network (DQN) architecture demonstrate that the proposed method with deep sparse representations from the multilayer sparse-coding structure can outperform existing sparse-coding-based DRL in control, for example, completing Mountain Car with 140.81 steps, achieving near 10% reward increase from the single-layer sparse-coding DRL algorithm, and obtaining 286.08 scores in Catcher, which are over two times the rewards of the other algorithms. Moreover, the proposed algorithm can reduce over 80% parameters while keeping performance improvements from deep sparse representations.

Inversion study of cadmium content in soil based on reflection spectroscopy and MSC-ELM model.

Heavy metal pollution in saline-alkali land has a significant impact on the ecological environment and human health. Rapid and accurate inversion of cadmium (Cd) element content in the saline-alkali land is important for environmental protection, saline-alkali soil improvement and conversion of saline-alkali land to cultivated land. Using traditional chemical detection methods to detect the content of heavy metal elements requires a long testing time and has the drawback of high prices. In this paper, we select the saline-alkali land of Zhenlai County as the study area and combine visible-NIR spectroscopy with machine learning models to invert the Cd content in the saline-alkali land. We preprocess the original reflection spectra using fractional order derivatives (FOD), then construct six three-band spectral indices (TBIs) and obtain the corresponding optimal band combination parameters by the optimal band combination (OBC) algorithm. To address the shortcomings of two-hidden-layer extreme learning machine (TELM), this paper introduces new weight parameters among the nodes of the first hidden layer, further extends it to multiple layers on this basis, and proposes the MSC-ELM model. The improved model is compared with several models, such as random forest (RF), partial least squares (PLS) and extreme learning machine (ELM). And the model performance is analyzed and compared by introducing several performance indicators, such as root mean square error (RMSE) and the ratio of the performance to interquartile (RPIQ). The experimental results show that the FOD transformation can eliminate the baseline drift and reduce the spectral noise. The constructed TBIs can effectively enhance the correlation with Cd content relative to the original single band, reduce redundant information and enhance the spectral features. The MSC-ELM model achieves better performance metrics compared to the other models and obtains the optimal prediction performance. This study provides an accurate and rapid method for the detection of Cd content in saline soil, which is important for the improvement and ecological recovery of saline soil.

Conformational and thermal properties of gluten in wheat dough as affected by bacterial cellulose.

Bacterial cellulose (BC), an important category of polysaccharides, was investigated as a texture improver in bakery products. This study focused on the changes in the conformational and thermal properties of gluten in the wheat dough system as affected by BC. Significant reductions in the free-SH content, fluorescence intensity, and surface hydrophobicity index (H0) were observed as a result of the increased BC addition. The electrophoresis profile (SDS-PAGE) and size exclusion (SE-HPLC) revealed the variation in molecular weight distribution, and the increase in the content of the 40-91 kDa molecular weight was at the expense of a decrease in the amount of the corresponding 10-40 kDa. When 0.1 % BC was added, both the α-helix and ß-sheet contents increased as a result of enhanced chemical interactions, thereby contributing to the gluten matrix with higher thermal stability. Further supplementation interfered with the current ordered gluten structure, which could be supported by the lower α-helix/ß-sheet content ratio and the decreased degradation temperature (Td) of gluten with 0.2 % BC. However, the observed decrease in the ratio of ß-turns to ß-sheets and weight loss at 600 °C indicated that a reconstructed gluten matrix induced by extra BC addition was formed to maintain the structural stability.

Coal Identification Based on Reflection Spectroscopy and Deep Learning: Paving the Way for Efficient Coal Combustion and Pyrolysis.

Coal plays an indispensable role in the world's energy structure. Coal converts chemical energy into energy such as electricity, heat, and internal energy through combustion. To realize the energy conversion of coal more efficiently, coal needs to be identified during the stages of mining, combustion, and pyrolysis. On this basis, different categories of coal are used according to industrial needs, or different pyrolysis processes are selected according to the category of coal. This paper proposes an approach combining deep learning with reflection spectroscopy for rapid coal identification in mining, combustion, and pyrolysis scenarios. First, spectral data of different coal samples were collected in the field and these spectral data were preprocessed. Then, an identification model combining a multiscale convolutional neural network (CNN) and an extreme learning machine (ELM), named RS_PSOTELM, is proposed. The effective features in the spectral data are extracted by the CNN, and the feature classification is realized utilizing the ELM. To enhance the identification performance of the model, we utilize a particle swarm optimization algorithm to optimize the parameters of the ELM. Experimental results show that RS_PSOTELM achieves 98.3% accuracy on the coal identification task and is able to identify coal quickly and accurately, providing a low-cost, efficient, and reliable approach for coal identification during the mining and application phases, as well as paving the way for efficient combustion and pyrolysis of coal.

Accelerated Log-Regularized Convolutional Transform Learning and Its Convergence Guarantee.

Convolutional transform learning (CTL), learning filters by minimizing the data fidelity loss function in an unsupervised way, is becoming very pervasive, resulting from keeping the best of both worlds: the benefit of unsupervised learning and the success of the convolutional neural network. There have been growing interests in developing efficient CTL algorithms. However, developing a convergent and accelerated CTL algorithm with accurate representations simultaneously with proper sparsity is an open problem. This article presents a new CTL framework with a log regularizer that can not only obtain accurate representations but also yield strong sparsity. To efficiently address our nonconvex composite optimization, we propose to employ the proximal difference of the convex algorithm (PDCA) which relies on decomposing the nonconvex regularizer into the difference of two convex parts and then optimizes the convex subproblems. Furthermore, we introduce the extrapolation technology to accelerate the algorithm, leading to a fast and efficient CTL algorithm. In particular, we provide a rigorous convergence analysis for the proposed algorithm under the accelerated PDCA. The experimental results demonstrate that the proposed algorithm can converge more stably to desirable solutions with lower approximation error and simultaneously with stronger sparsity and, thus, learn filters efficiently. Meanwhile, the convergence speed is faster than the existing CTL algorithms.

Optimal BR signalling is required for adequate cell wall orientation in the Arabidopsis root meristem.

Plant brassinosteroid hormones (BRs) regulate growth in part through altering the properties of the cell wall, the extracellular matrix of plant cells. Conversely, feedback signalling from the wall connects the state of cell wall homeostasis to the BR receptor complex and modulates BR activity. Here, we report that both pectin-triggered cell wall signalling and impaired BR signalling result in altered cell wall orientation in the Arabidopsis root meristem. Furthermore, both depletion of endogenous BRs and exogenous supply of BRs triggered these defects. Cell wall signalling-induced alterations in the orientation of newly placed walls appear to occur late during cytokinesis, after initial positioning of the cortical division zone. Tissue-specific perturbations of BR signalling revealed that the cellular malfunction is unrelated to previously described whole organ growth defects. Thus, tissue type separates the pleiotropic effects of cell wall/BR signals and highlights their importance during cell wall placement.

Electro-mechanical coupling diode of elastic wave in nonlinear piezoelectric metamaterials.

In this investigation, the bandgaps and nonreciprocal transmission of the nonlinear piezoelectric phononic crystal and elastic wave metamaterial are studied. Analytical solutions for the wave motion equations with the electro-mechanical coupling are obtained. According to the continuous conditions, the stop bands and transmission coefficients of both fundamental wave and second harmonic are derived by the stiffness matrix method. Some particular examples are presented to show the nonreciprocal transmission of the nonlinear elastic waves. Additionally, nonlinear ultrasonic experiments are applied to verify the theoretical analyses and numerical simulations. This work is intended to be helpful in the design and fabrication of devices of the elastic wave diode with piezoelectric materials.

A Randomized, Double-Blind, Single-Dose Study Comparing the Biosimilarity of HOT-1010 With Bevacizumab (Avastin®) in Chinese Healthy Male Subjects.

Objective: This study was conducted to compare the pharmacokinetics, safety and immunogenicity of HOT-1010 with bevacizumab (Avastin®) in Chinese healthy male subjects. Methods: A single-center, randomized, double-blind, single-dose, parallel trial was performed in 84 Chinese healthy male subjects who randomly (1:1) received a single intravenous infusion of 1 mg/kg HOT-1010 or Avastin® for 90 min and followed up for 85 days. Serum concentrations of bevacizumab were analyzed by enzyme-linked immunosorbent assay. Primary pharmacokinetic parameters, Cmax, AUC0-t and AUC0-∞, were calculated and evaluated the bioequivalence between HOT-1010 and Avastin®, the safety and immunogenicity of investigational drugs were also assessed. Results: A total of 82 subjects completed the study. The 90% Confidence Intervals for geometric mean ratios of Cmax, AUC0-t and AUC0-∞ were 91.81-103.64%, 85.19-95.39% and 85.04-95.36%, which were all within the bioequivalence margin. Treatment-emergent adverse events were reported in 27 (65.9%) subjects in HOT-1010 group and 23 (56.1%) subjects in Avastin® group. Most TEAEs were mild or moderate. No TEAEs, Serious Adverse Events or deaths leading to discontinuation was reported. Subjects were all tested negative for Anti-drug Antibody. Conclusion: HOT-1010 exhibited the similar pharmacokinetics, safety and immunogenicity profiles of bevacizumab (Avastin®) in Chinese healthy male subjects. Clinical Trial Registration: http://www.chinadrugtrials.org.cn/index.html, CTR20181610.

A Residual Network and FPGA Based Real-Time Depth Map Enhancement System.

Depth maps obtained through sensors are often unsatisfactory because of their low-resolution and noise interference. In this paper, we propose a real-time depth map enhancement system based on a residual network which uses dual channels to process depth maps and intensity maps respectively and cancels the preprocessing process, and the algorithm proposed can achieve real-time processing speed at more than 30 fps. Furthermore, the FPGA design and implementation for depth sensing is also introduced. In this FPGA design, intensity image and depth image are captured by the dual-camera synchronous acquisition system as the input of neural network. Experiments on various depth map restoration shows our algorithms has better performance than existing LRMC, DE-CNN and DDTF algorithms on standard datasets and has a better depth map super-resolution, and our FPGA completed the test of the system to ensure that the data throughput of the USB 3.0 interface of the acquisition system is stable at 226 Mbps, and support dual-camera to work at full speed, that is, 54 fps@ (1280 × 960 + 328 × 248 × 3).

Tunable mechanical diode of nonlinear elastic metamaterials induced by imperfect interface.

In this investigation, the non-reciprocal transmission in a nonlinear elastic metamaterial with imperfect interfaces is studied. Based on the Bloch theorem and stiffness matrix method, the band gaps and transmission coefficients with imperfect interfaces are obtained for the fundamental and double frequency cases. The interfacial influences on the transmission behaviour are discussed for both the nonlinear phononic crystal and elastic metamaterial. Numerical results for the imperfect interface structure are compared with those for the perfect one. Furthermore, experiments are performed to support the theoretical analysis. The present research is expected to be helpful to design tunable devices with the non-reciprocal transmission and diode behaviour of the elastic metamaterial.