Parallel Extraction Research Articles

Computer motherboard assembly defect detection using parallel feature extraction and progressive feature fusion

Computer motherboard assembly defect detection using parallel feature extraction and progressive feature fusion

SWFormer: Stochastic Windows Convolutional Transformer for Hybrid Modality Hyperspectral Classification.

Joint classification of hyperspectral images with hybrid modality can significantly enhance interpretation potentials, particularly when elevation information from the LiDAR sensor is integrated for outstanding performance. Recently, the transformer architecture was introduced to the HSI and LiDAR classification task, which has been verified as highly efficient. However, the existing naive transformer architectures suffer from two main drawbacks: 1) Inadequacy extraction for local spatial information and multi-scale information from HSI simultaneously. 2) The matrix calculation in the transformer consumes vast amounts of computing power. In this paper, we propose a novel Stochastic Window Transformer (SWFormer) framework to resolve these issues. First, the effective spatial and spectral feature projection networks are built independently based on hybrid-modal heterogeneous data composition using parallel feature extraction, which is conducive to excavating the perceptual features more representative along different dimensions. Furthermore, to construct local-global nonlinear feature maps more flexibly, we implement multi-scale strip convolution coupled with a transformer strategy. Moreover, in an innovative random window transformer structure, features are randomly masked to achieve sparse window pruning, alleviating the problem of information density redundancy, and reducing the parameters required for intensive attention. Finally, we designed a plug-and-play feature aggregation module that adapts domain offset between modal features adaptively to minimize semantic gaps between them and enhance the representational ability of the fusion feature. Three fiducial datasets demonstrate the effectiveness of the SWFormer in determining classification results.

Dominating Alzheimer's disease diagnosis with deep learning on sMRI and DTI-MD.

Alzheimer's disease (AD) is a progressive and irreversible neurodegenerative disorder that has become one of the major health concerns for the elderly. Computer-aided AD diagnosis can assist doctors in quickly and accurately determining patients' severity and affected regions. In this paper, we propose a method called MADNet for computer-aided AD diagnosis using multimodal datasets. The method selects ResNet-10 as the backbone network, with dual-branch parallel extraction of discriminative features for AD classification. It incorporates long-range dependencies modeling using attention scores in the decision-making layer and fuses the features based on their importance across modalities. To validate the effectiveness of our proposed multimodal classification method, we construct a multimodal dataset based on the publicly available ADNI dataset and a collected XWNI dataset, which includes examples of AD, Mild Cognitive Impairment (MCI), and Cognitively Normal (CN). On this dataset, we conduct binary classification experiments of AD vs. CN and MCI vs. CN, and demonstrate that our proposed method outperforms other traditional single-modal deep learning models. Furthermore, this conclusion also confirms the necessity of using multimodal sMRI and DTI data for computer-aided AD diagnosis, as these two modalities complement and convey information to each other. We visualize the feature maps extracted by MADNet using Grad-CAM, generating heatmaps that guide doctors' attention to important regions in patients' sMRI, which play a crucial role in the development of AD, establishing trust between human experts and machine learning models. We propose a simple yet effective multimodal deep convolutional neural network model MADNet that outperforms traditional deep learning methods that use a single-modality dataset for AD diagnosis.

An improved YOLOv7 network using RGB-D multi-modal feature fusion for tea shoots detection

Due to the increasing scarcity of tea pickers, the implementation of intelligent harvesting for premium tea is a crucial prerequisite for the sustainable development of the premium tea industry. The initial step towards achieving intelligent and precise harvesting is the accurate detection of tender shoots, which consist of one bud and one leaf. However, accurately identifying tea shoots poses a challenging visual task due to their small size, variable shapes, as well as similar colors and backgrounds. The existing model, based on RGB images, can only detect partial targets. To address this issue and further enhance the detection of tea buds, this study proposes the utilization of multi-modal features encompassing red, green, blue, and depth (RGB-D) for identification. In addition, a unidirectional complementary multi-modal fusion method is introduced to minimize the adverse effects caused by low-quality depth information. Firstly, an RGB-D dataset comprising high-quality tea leaves is constructed, and the samples are carefully calibrated. Subsequently, an enhanced end-to-end RGB-D multi-modal object detection network, referred to as YOLO-RGBDtea, is developed based on You Only Look Once version 7 (YOLOv7). This model incorporates a parallel lightweight depth image feature extraction backbone network and incorporates a self-attention mechanism to prioritize contextual information. Lastly, a cross-modal spatial attention fusion module (CSFM) is devised to collaboratively integrate depth features with RGB features in a unidirectional manner. The experimental results reveal that YOLO-RGBDtea achieves an AP50 of 91.12% when confronting complex outdoor tea shoots, exhibiting significant performance improvements compared to YOLOv7, especially in scenarios involving small targets, overlapping target groups, and highly overexposed images. Notably, the parameter increment in YOLO-RGBDtea compared to the original YOLOv7 model is merely 17.8%, and the additional components can be seamlessly transferred to other models. Overall, this study introduces a straightforward yet effective multi-modal fusion method that bears theoretical and practical significance in advancing the detection of high-quality tea shoots in complex outdoor environments.

Medium‐term load forecasting of power system based on BiLSTM and parallel feature extraction network

AbstractWith the diversification of users’ energy demands, accurate load forecasting is an important prerequisite for optimal scheduling and economic operation of the system, but a single‐load forecasting method cannot effectively predict multi‐energy loads accurately. Therefore, this paper proposes a multi‐energy load forecasting method based on bidirectional long short‐term memory (BiLSTM) and parallel feature extraction networks. Firstly, residual network and convolutional block attention module were used to extract the spatial coupling features of multi‐energy load data. Secondly, BiLSTM is used to capture the temporal features and long‐term dependencies in the load data, and the spatial coupling features are fused to obtain non‐linear prediction results. Finally, the non‐linear prediction results and the linear prediction results obtained by using multi‐energy linear regression were linearly superimposed to obtain the final prediction results. In this paper, IES load data of Tempe Campus of Arizona State University was used to verify and compared with several existing methods, and the results showed that Weighted Mean Absolute Percentage Error decreased by more than 20%.

Parallel artificial liquid membrane extraction coupled with UPLC-ESI-MS/MS method for high-throughput quantitation of repaglinide in diabetic patients

A high-throughput therapeutic monitoring method was developed for repaglinide (RPG) in diabetic patients, combining parallel artificial liquid membrane extraction (PALME) with ultraperformance liquid chromatography electrospray ionization tandem mass spectrometry (UPLC-ESI-MS/MS). PALME was performed using a 96-well donor plate comprising a donor solution containing a plasma sample, 50 mM phosphate buffer (pH = 8.0), and cetirizine (CTZ) as internal standard. A polypropylene (PP) porous membrane served as a selective support for the liquid membrane (SLM), preventing nonspecific binding produced by other membranes. The extraction was accomplished across SLM made of PP membrane with dodecyl acetate and 1 % trioctylamine (w/w), and the acceptor solution comprised DMSO and 200 mM formic acid (50:50, v/v). The simple workflow for PALME provided analyte enrichment, highly efficient sample cleanup, high throughput analysis, and excellent reproducibility. Method validation met FDA criteria, with a linear plasma calibration range (0.1–100 ng mL−1, r = 0.9995) and a lower limit of quantitation (LLOQ) of 0.1 ng mL−1. Recovery results at 98.9 % affirmed method reliability. The ability to analyze 198 samples per hour, coupled with a reduced amount of solvents, underscores the method's high throughput and eco-friendly profile. The PALME-UPLC-ESI-MS/MS method was successfully applied to therapeutic drug monitoring of RPG in diabetic patients following 2 mg RPG tablet administration, establishing its effectiveness.

Concatenation-based pre-trained convolutional neural networks using attention mechanism for environmental sound classification

Environmental sound classification (ESC) is gaining popularity in the field of information processing due to its significance in non-speech audio categorization. ESC faces challenges in categorizing ambient sounds, which lack a clear structure, unlike speech or music sounds. Deep learning (DL) techniques are widely used in ESC to extract relevant information from ambient sounds, in which feature extraction is very crucial and directly impacts the classification performance. However, feature extraction can be computationally expensive, especially when employing complex non-linear techniques, due to the significant amount of computing resources required during the training process of DL models. Moreover, environmental sounds can exhibit significant variability in terms of their temporal and spectral characteristics, which can pose challenges to training DL models effectively. In order to overcome these limitations, this research proposes an efficient method for extracting meaningful features from audio files and improving DL techniques' performance using spectrogram images generated from sound environmental datasets. The proposed approach uses convolutional neural networks (CNNs) with attention mechanisms and appropriate data augmentation methods. Unlike pre-trained models that use a single vector for feature extraction, the proposed approach uses a new concatenation-based CNN model with attention mechanisms, which can more effectively capture intricate relationships between input data features. This approach allows for the extraction of features from different sectors of the feature space, resulting in a more precise classification of complex and diverse datasets. Additionally, the proposed approach leverages the parallel extraction feature technique to extract features from multiple CNN models, which improves classification performance. Furthermore, attention modules are used to focus on the most relevant features of the input data. Comparative experiments are also conducted for the proposed approach and some existing state-of-the-art methods, and the results show that the former has better classification performance.

Antibacterial Activity and Untargeted Metabolomics Profiling of Acalypha arvensis Poepp.

The search for potent antimicrobial compounds is critical in the face of growing antibiotic resistance. This study explores Acalypha arvensis Poepp. (A. arvensis), a Caribbean plant traditionally used for disease treatment. The dried plant powder was subjected to successive extractions using different solvents: hexane (F1), dichloromethane (F2), methanol (F3), a 50:50 mixture of methanol and water (F4), and water (F5). Additionally, a parallel extraction was conducted using a 50:50 mixture of methanol and chloroform (F6). All the fractions were evaluated for their antimicrobial activity, and the F6 fraction was characterized using untargeted metabolomics using SPME-GC×GC-TOFMS. The extracts of A. arvensis F3, F4, and F5 showed antibacterial activity against Staphylococcus aureus ATCC 25923 (5 mg/mL), MRSA BA22038 (5 mg/mL), and Pseudomonas aeruginosa ATCC 27853 (10 mg/mL), and fraction F6 showed antibacterial activity against Staphylococcus aureus ATCC 29213 (2 mg/mL), Escherichia coli ATCC 25922 (20 mg/mL), Pseudomonas aeruginosa ATCC 27853 (10 mg/mL), Enterococcus faecalis ATCC 29212 (10 mg/mL), Staphylococcus aureus 024 (2 mg/mL), and Staphylococcus aureus 003 (2 mg/mL). Metabolomic analysis of F6 revealed 2861 peaks with 58 identified compounds through SPME and 3654 peaks with 29 identified compounds through derivatization. The compounds included methyl ester fatty acids, ethyl ester fatty acids, terpenes, ketones, sugars, amino acids, and fatty acids. This study represents the first exploration of A. arvensis metabolomics and its antimicrobial potential, providing valuable insights for plant classification, phytochemical research, and drug discovery.

Unveiling threats to glacier biota: Bioaccumulation, mobility, and interactions of radioisotopes with key biological components

Contemporary melting glaciers are considered a secondary source of pollutants including radionuclides. Cryoconite - biogenic sediment on the glacier surface - exhibits high concentrations of natural and anthrophogenic radioisotopes. Understanding the interactions between radioisotopes and organisms is essential for evaluating their potential impact on glacier-related ecosystems. Using a multidimensional approach and intensive sampling (17 glaciers), we investigated the relationships between activity concentrations of 137Cs and 210Pb and various biotic components of cryoconite such as the amount of organic matter, chlorophyll concentration, the ratio of cyanobacteria to all bacteria, and size of cryoconite granules. Additionally, to better understand the bioavailability and fate of radioisotopes in this ecosystem, we measured the uptake ratio of 137Cs, 210Pb, 238Pu, and 239+240Pu in the top consumers, and examined the mobility of radioisotopes by measuring 137Cs and 210Pb activity concentrations after a parallel extraction using media with different specific ion exchange capacities. The activity concentrations of both 137Cs and 210Pb showed a large variability reaching 5.8 kBq kg−1 and 7.2 kBq kg−1, respectively. Their activity concentrations were positively related to the amount of organic matter, however, the 210Pb also increased with the chlorophyll concentration. This might be due to the difference in the deposition of both elements: lead, being deposited constantly, binds also to currently developing communities, while caesium deposition peaked in the 1960s. The mobility analysis revealed that the loosely bound fraction of 210Pb was more strongly related to organic-metallic complexes than 137Cs. Firmly bound radioisotope fractions (anhydrous interlayer sites of minerals) were three times higher for 137Cs than for 210Pb. The median uptake ratios of radioisotopes were determined as 0.07–0.111 for 137Cs, 0.177 for 210Pb, 0.07 for 239+240Pu. Our findings emphasize the importance of organisms in the accumulation of radioisotopes on glaciers and suggest an impact of radionuclides on glacier organisms.

Incorporating stock prices and text for stock movement prediction based on information fusion

Forecasting stock market via historical financial data is an important issue for market participants because even if the prediction accuracy is only slightly improved, better trading decisions can be made. Historical financial data has evolved from the initial single text or stock price to the fusion of multisource information. However, how to adopt a method that adaptively fuses numerical data and text so that the prediction model can learn time series information in parallel remains a challenging problem. In this paper, we propose a collaborative attention Transformer fusion model for stock movement prediction (CoATSMP), including parallel extraction of text and prices features, parameter-level fusion and a joint feature processing module, that can successfully deeply fuse text and stock prices in view of the soft fusion method. The experiments show that (1) the proposed approach outperforms the baselines, (2) the soft fusion method proposed in this paper has better modeling performance under the CoATSMP framework, which brings greater improvement in the prediction performance, (3) models containing prices and text are better than those using only one data source, and (4) quantitative analysis of experimental results indicates that text plays a relatively more critical role in the CoATSMP framework. Real simulation trading shows that the trading strategy based on CoATSMP can significantly improve profits; thus, the model has practical application value.

Determination of microplastics by FTIR spectroscopy based on quaternion parallel feature fusion and support vector machine

This study proposed a FTIR spectroscopy combing with quaternion parallel feature extraction methodology for the first time to address the qualitative and quantitative analysis of microplastics. For the international available microplastic FTIR dataset, the original, the first derivative and the second derivative FTIR spectra of polyethylene(PE), PE + fouling and polypropylene(PP) samples were first formed as an integral representation by a pure quaternion matrix. Then quaternion principal component analysis(QPCA) was conducted to obtain quaternion feature vectors and support vector machine(SVM) was followed to establish the qualitative model. Meanwhile, we experimentally mixed four types of microplastics of polyvinyl chloride(PVC), polystyrene (PS), PP and PE into sand to prepare simulated samples and collected their FTIR spectra. The concentrations of microplastics were 0.5%, 1%, 1.5%, 3%, 5%, 10%, respectively. The QPCA based parallel feature extraction combing with SVR was conducted to establish the quantitative model. The results show that QPCA has transcended traditional PCA and serial feature extraction methods. Furthermore, compared with traditional algorithms of PLS, BPNN, RF, and 1D-CNN, QPCA-SVM has achieved the highest values of accuracy, precision, recall and F1-score, which were 0.98571, 1, 0.99412 and 0.99697, respectively. QPCA-SVR has achieved the smallest RMSEP for four kinds of microplastics, which were 0.26361, 0.80558, 0.37099, and 0.16246, respectively. This work demonstrates QPCA fuses more spectral characteristics and reduces the redundancy of the complex vector, which is beneficial to improve the recognition and regression ability of SVM. The “QPCA enabled FTIR spectroscopy technique” in this study provides a new idea to promote the rapid and accurate determination of microplastics.

Novel imbalanced subdomain adaption multiscale convolutional network for cross-domain unsupervised fault diagnosis of rolling bearings

Data on the vibration signals collected from rolling bearings mostly belongs to health conditions, leading to an imbalanced data distribution. In addition, frequent switching of operating conditions results in unlabeled data collected under a specific working condition. This paper proposes a novel network for cross-domain unsupervised fault diagnosis of rolling bearings considering the imbalanced data to address these challenges. First, a multiscale parallel features extraction is developed, which can fully mine the rich high-level feature representation of various fault types from the original data and has a high value for fault identification. Second, a squeeze-and-excitation attention mechanism is constructed to enhance features conducive to model classification and suppress redundant features. Finally, a new loss function is proposed to optimize the model, which can accurately classify imbalanced source domain and easily align related subdomains of two domains. The proposed method was validated on multiple unsupervised cross-domain diagnostic tasks on two bearing datasets. Experimental results manifest that the proposed method has stable generalization performance and excellent robustness.

A power quality disturbances classification method based on multi-modal parallel feature extraction

Power quality disturbance (PQD) is an important problem affecting the safe and stable operation of power system. Traditional single modal methods not only have a large number of parameters, but also usually focus on only one type of feature, resulting in incomplete information about the extracted features, and it is difficult to identify complex and diverse PQD types in modern power systems. In this regard, this paper proposes a multi-modal parallel feature extraction and classification model. The model pays attention to both temporal and spatial features of PQD, which effectively improves classification accuracy. And a lightweight approach is adopted to reduce the number of parameters of the model. The model uses Long Short Term Memory Neural Network (LSTM) to extract the temporal features of one-dimensional temporal modes of PQD. At the same time, a lightweight residual network (LResNet) is designed to extract the spatial features of the two-dimensional image modality of PQD. Then, the two types of features are fused into multi-modal spatio-temporal features (MSTF). Finally, MSTF is input to a Support Vector Machine (SVM) for classification. Simulation results of 20 PQD signals show that the classification accuracy of the multi-modal model proposed in this paper reaches 99.94%, and the parameter quantity is only 0.08 MB. Compared with ResNet18, the accuracy of the proposed method has been improved by 2.55% and the number of parameters has been reduced by 99.25%.

Tertiary drug information sources for treatment and prevention of COVID-19.

To evaluate tertiary drug information databases in terms of scope, consistency of content, and completeness of COVID-19 drug information. Five electronic drug information databases: Clinical Pharmacology, Lexi-Drugs, AHFS DI (American Hospital Formulary Service Drug Information), eFacts and Comparisons, and Micromedex In-Depth Answers, were evaluated in this cross-sectional evaluation study, with data gathered from October 2021 through February 2022. Two study investigators independently extracted data (parallel extraction) from each resource. Descriptive statistics were primarily used to evaluate scope (i.e., whether the resource addresses use of the medication for treatment or prevention of COVID-19) and completeness of content (i.e., whether full information is provided related to the use of the medication for treatment or prevention of COVID-19) based on a 10-point scale. To analyze consistency among resources for scope, the Fleiss multi-rater kappa was used. To analyze consistency among resources for type of recommendation (i.e., in favor, insufficient evidence, against), a two-way mixed effects intraclass coefficient was calculated. A total of 46 drug monographs, including 3 vaccination monographs, were evaluated. Use of the agents for treatment of COVID-19 was most frequently addressed in Lexi-Drugs (73.9%), followed by eFacts and Comparisons (71.7%), and Micromedex (54.3%). The highest overall median completeness score was held by AHFS DI followed by Micromedex, and Clinical Pharmacology. There was moderate consistency in terms of scope (kappa 0.490, 95% CI 0.399-0.581, p<0.001) and recommendations (intraclass correlation coefficient 0.518, 95% CI 0.385-0.651, p<0.001). Scope and completeness results varied by resource, with moderate consistency of content among resources.

SSCT-Net: A Semisupervised Circular Teacher Network for Defect Detection With Limited Labeled Multiview MFL Samples

Deep learning methods have demonstrated promising performance in magnetic flux leakage (MFL) defect detection under adequate amounts of labeled samples. However, in industrial occasions, obtaining adequate amounts of labeled samples is time-consuming and expensive, and applying only limited labeled samples can lead to unsatisfactory defect detection accuracy. To address the above issues, a defect detection method named semi-supervised circular teacher network (SSCT-Net) is proposed in this article. First, a parallel feature extraction network with hybrid attention is proposed in SSCT-Net so that the useful features of multi-view MFL signals can be extracted simultaneously. Second, semi-supervised circular learning is proposed for the first time. In semi-supervised circular learning, a distinguishable feature embedding space is constructed, and two structurally identical deep networks co-supervise and collaborate through the proposed consistent circular strategy so that the decision bias of unlabeled samples can be reduced. Finally, the trained model is applied for defect detection in practice. The proposed method can establish a potential connection between multi-view MFL signals and fully utilize labeled and unlabeled MFL signals. The experiments in simulations and real-world applications demonstrate that SSCT-Net can reach 92% detection accuracy with only 20% labeled samples, which is more effective than the state-of-the-art methods and leads to a promising practical utility of the proposed method.

Multiscale global and local self-attention-based network for remaining useful life prediction

Remaining useful life (RUL) prediction plays an important role in prognostics and health management (PHM) and can significantly enhance equipment reliability and safety in various engineering applications. Accurate RUL prediction enables proactive maintenance planning, helping prevent potential hazards and economic losses caused by equipment failures. Recently, while deep learning-based methods have swept the RUL prediction field, most existing methods still have difficulties in simultaneously extracting multiscale global and local degradation feature information from raw multi-sensor monitoring data. To address these issues, we propose a novel multiscale global and local self-attention-based network (MGLSN) for RUL prediction. MGLSN consists of global and local feature extraction subnetworks based on self-attention, which work in parallel to simultaneously extract the global and local degradation features of equipment and can adaptively focus on more important parts. While the global network captures long-term dependencies between time steps, the local network focuses on modeling local temporal dynamics. The design of parallel feature extraction can avoid the mutual influence of information from global and local aspects. Moreover, MGLSN adopts a multiscale feature extraction design (multiscale self-attention and convolution) to capture the global and local degradation patterns at different scales, which can be combined to better reflect the degradation trend. Experiments on the widely used Commercial Modular Aero-Propulsion System Simulation (CMAPSS), New CMAPSS (N-CMAPSS), and International Conference on Prognostics and Health Management 2008 challenge datasets provided by NASA show that MGLSN significantly outperforms state-of-the-art RUL prediction methods and has great application prospects in the field of PHM.

Bearing fault diagnosis based on CNN-BiLSTM and residual module

Bearings are key components of rotating machinery, and their fault diagnosis is essential for machinery operation. Bearing vibration signals belong to time series data, but traditional convolutional neural networks (CNNs) or recurrent neural networks cannot fully extract the fault features from these signals. To address the insufficient feature extraction and poor noise resistance, this paper proposes a fault diagnosis model based on continuous wavelet transform (CWT), CNN with channel attention, bidirectional long short-term memory network (BiLSTM) and residual module. Firstly, a parallel dual-path feature extraction mechanism is constructed which takes time-domain signals and time–frequency images transformed via CWT as the input respectively. Then BiLSTM extracts the time features of the signal as one path, and the CNN with efficient channel attention extracts the spatial features as the other path. This parallel neural network contributes to better feature extraction. Then, the residual module is applied to extract the global features to further improve the feature extraction ability and noise immunity. The experimental results demonstrate that the proposed model on the Case Western Reserve University dataset has better diagnostic accuracy under different working conditions and different signal-to-noise ratios than other methods. In addition, the model shows good generalization performance on Jiangnan University dataset.

TranSiam: Aggregating multi-modal visual features with locality for medical image segmentation

Automatic segmentation of medical images plays an important role in the diagnosis of diseases. On single-modal data, convolutional neural networks have demonstrated satisfactory performance. However, multi-modal data encompasses a greater amount of information rather than single-modal data. Multi-modal data can be effectively used to improve the segmentation accuracy of regions of interest by analyzing both spatial and temporal information. In this study, we propose a dual-path segmentation model for multi-modal medical images, named TranSiam. Taking into account that there is a significant diversity between the different modalities, TranSiam employs two parallel CNNs to extract the features which are specific to each of the modalities. In our method, two parallel CNNs extract detailed and local information in the low-level layer, and the Transformer layer extracts global information in the high-level layer. Finally, we fuse the features of different modalities via a locality-aware aggregation block (LAA block) to establish the association between different modal features. The LAA block is used to locate the region of interest and suppress the influence of invalid regions on multi-modal feature fusion. TranSiam uses LAA blocks at each layer of the encoder in order to fully fuse multi-modal information at different scales. Extensive experiments on several multi-modal datasets have shown that TranSiam achieves satisfying results.

Predicting Forex Currency Fluctuations Using a Novel Bio-Inspired Modular Neural Network

In the realm of foreign exchange (Forex) market predictions, Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have been commonly employed. However, these models often exhibit instability due to vulnerability to data perturbations attributed to their monolithic architecture. Hence, this study proposes a novel neuroscience-informed modular network that harnesses closing prices and sentiments from Yahoo Finance and Twitter APIs. Compared to monolithic methods, the objective is to advance the effectiveness of predicting price fluctuations in Euro to British Pound Sterling (EUR/GBP). The proposed model offers a unique methodology based on a reinvigorated modular CNN, replacing pooling layers with orthogonal kernel initialisation RNNs coupled with Monte Carlo Dropout (MCoRNNMCD). It integrates two pivotal modules: a convolutional simple RNN and a convolutional Gated Recurrent Unit (GRU). These modules incorporate orthogonal kernel initialisation and Monte Carlo Dropout techniques to mitigate overfitting, assessing each module’s uncertainty. The synthesis of these parallel feature extraction modules culminates in a three-layer Artificial Neural Network (ANN) decision-making module. Established on objective metrics like the Mean Square Error (MSE), rigorous evaluation underscores the proposed MCoRNNMCD–ANN’s exceptional performance. MCoRNNMCD–ANN surpasses single CNNs, LSTMs, GRUs, and the state-of-the-art hybrid BiCuDNNLSTM, CLSTM, CNN–LSTM, and LSTM–GRU in predicting hourly EUR/GBP closing price fluctuations.

Multi-scale coal and gangue detection in dense state based on improved Mask RCNN

Coal and gangue intelligent sorting accounts for a vital proportion of modern coal mines, which takes on critical significance in boosting clean coal utilization. Aiming at the problem of low recognition accuracy that arises from the large diversity in particle size of raw coal and the adhesion and half-occlusion between them, A Mask RCNN-based instance segmentation network for coal and gangue image is proposed. First, multi-branch parallel feature extraction bottlenecks are developed to build a lightweight backbone with a mixed attention mechanism and self-correcting convolution incorporated to enhance the feature representation of targets. Subsequently, a necknet is developed, aggregating the context feature information of the backbone from channel and space to enhance the position and boundary information of targets. Lastly, the effectiveness of this method is fully validated by ablation and test experiments using self-built coal and gangue RGB image datasets include varieties of characteristics as input. As indicated by the experimental results, the above-described method is capable of enhancing the segmentation ability of stacked and adhered coal and gangue and reducing the missed detection rate of small targets. ISNet_CG achieves mAP, mIoU, F1-scores, and FPS of 98.0, 96.5, 0.987, and 24.2 FPS, respectively, such that the correct category and location can be provided for coal and gangue in a dense state.