Search

Accurate characterization is an important issue in paper currency recognition system. This paper proposes a robust paper currency recognition method based on Hidden Markov Model (HMM). By employing HMM, the texture characteristics of paper currencies are modeled as a random process. The proposed algorithm can be used for distinguishing paper currency from different countries. A similarity measure has been used for the classification in the proposed algorithm. To evaluate the performance of the proposed algorithm, experiments have been conducted on more than 100 denominations from different countries. The results indicate 98% accuracy for recognition of paper currency. [Copyright &y& Elsevier]

The paper recommender system has provided great convenience for people to obtain knowledge of what they are interested in. Existing recommendation works mainly rely on learning entities' feature representations (network embedding) from the Heterogeneous Information Networks (HINs) through meta-paths. Most of them suffer from the difficulty of designing and selecting meta-paths in HIN. Moreover, changes in papers' captive audiences and users' reading favor in the recent trends are always covered up by long-term trends. To this end, we introduce a multi-attention aware recommendation method, named MARec. Specifically, to efficiently design and select meta-paths, we propose an auxiliary node-driven HIN construction method and combine it with the graph attention network (GAT) to learn the entities' feature representations. To endow the recommendation results with timeliness, we design an attention-aware Bi-directional LSTM (Bi-LSTM) and a compensation mechanism to incorporate the recent trend changes. Furthermore, to take advantage of the importance of different papers/users, we utilize an attention mechanism to capture their importance by adaptively fusing the interactive historical data of those paper/user entities. We use extensive experiments on three real-world datasets to show that our proposed MARec outperforms current representative baselines by 11.49%, 11.40%, and 13.37% on average on three main metrics. • A general heterogeneous information network towards simplifying meta-path design. • Embedding with attention mechanism avoids painstakingly selecting meta-paths. • Feature extraction under multi-time scale greatly improves the recommended results. • MARec demonstrates superior performance on different real-world domain datasets. [ABSTRACT FROM AUTHOR]

Long time series forecasting is widely used in areas such as power dispatch, traffic control, and weather forecasting. The pattern of seasonality and trends in long time series are often complex, especially when they are presented at different time scales. Existing methods typically focus on only one scale or randomly select scales, which leads to a significant loss of valuable information. Additionally, current methods often transform multi-channel data into a single-channel format, ignoring interactions and complex relationships between channels. The paper proposes MultiWaveNet, a novel long time series forecasting framework that addresses seasonality as well as trends separately. For the seasonal component, the framework uses multi-scale wavelet decomposition to generate subseries at multiple scales. A learnable optimization factor is introduced simultaneously to separate high-frequency components mixed in low-frequency series after wavelet decomposition. In order to reduce information redundancy and model complexity, the paper develops a wavelet domain sampling encoder that consists of just one Transformer encoder, ensuring effective modeling of long-term dependencies while maintaining feature extraction effectiveness. As for the trend component, unlike previous research, the weights of channels are adjusted based on their importance, allowing the more crucial channels to have a greater impact and thereby addressing the limitations of individual processing methods. The paper performs extensive experiments on nine standard datasets, demonstrating that MultiWaveNet is the most competitive method. • Learnable optimization factor eliminate high-frequency information mixed in the low-frequency subseries. • Developed a wavelet-domain sampling encoder incorporating only the Transformer Encoder. • An attention-based channel-aware module was designed. [ABSTRACT FROM AUTHOR]

The detection of epileptic seizures is key for neurologists to initiate the right treatment at the earliest. However, the traditional methods are dependent on manual diagnosis which are error-prone. Hence, the current article presents an automated machine learning-based approach with competing results to detect seizures and its types using traditional features. Towards this goal, the present paper explores the performance of two fractal parameters namely Higuchi and, Kat'z fractal dimension, along with power spectral density and spectral entropy to detect EEG signals with seizure and non-seizure characteristics. To validate the performance via cross-validation technique, the features are extracted and experimented from three publically available databases such as Bern-Barcelona (focal, non-focal), Khas (preictal, interictal, ictal), and Bonn. Thereafter evaluated with different machine learning algorithms like Random Forest classifier, Ada Boost classifier, Gradient Boosting classifier, Extra Tree classifier, SVM classifier, and Naive Bayes classifier. The experiments showed 100% F1 score and accuracy in classifying focal and non-focal signals for random forest and extra tree classifier methods. To the best of our knowledge, this is the first paper reporting this score for the Bern-Barcelona dataset. The same features are also experimented for the detection of interictal, ictal and preictal scenarios. In the process, 100% accuracy is obtained in classifying interictal and ictal EEG signals, and an accuracy of 94% is attained in classifying interictal-preictal signals. This score is 14% higher than the current state-of-the-art methods using the same database. The features are also used in the Bonn database, where the proposed approach gave an accuracy of 100% same as the state-of-the-art methods. • Obtained competing evaluation score relative to SOTA methods using ML methods. • Competing score reported for interictal-preictal based classification of EEG data. • EEG classification is performed in the sensing time domain itself. • The proposed approach is validated using three benchmark datasets. • The performance is evaluated using cross-validation technique. [ABSTRACT FROM AUTHOR]

The emergence of artificial intelligence (AI) and its related capabilities has led industries to rethink the existing practices of conventional supply chain management and data analysis. Machine learning (ML), Deep Learning (DL) and their unique ability to predict future data and classify data have led to important research in the supply chain (SC) domain, particularly in identifying and prioritizing supply chain risks. This paper proposes several DL methodologies to exploit the benefit of DL, particularly to identify whether any product will be delivered late due to any unforeseen reason in a complex SC system. Four different DL architectures (Simple-LSTM, Deep-LSTM, 1D-CNN, and TCN-1DSPCNN models) are proposed to extract features, while six variant classifiers: Softmax, random trees (RT), random forest (RF), K-nearest neighbor (KNN), artificial neural network (ANN), and support vector machine (SVM), were used to classify delay or non-delay information. By seamlessly capturing intricate temporal dependencies, these DL models enhance accuracy in robustly identifying supply chain late orders. Leveraging their hierarchical feature learning, these proposed DL models excel in recognizing subtle patterns and correlations, making them ideal for classifying late orders within the supply chain. Their parallel processing prowess facilitates real-time decision support, allowing organizations to address potential delays and allocate resources effectively and proactively. Five-fold cross-validation is presented to avoid over-fitting and to prove the efficiency of the proposed DL models. The total accuracies of the six ML classifiers are 74.03, 75.81, 93.35, 87.72, 93.59, and 95.10, respectively, while the maximum accuracies obtained from four proposed DL methodologies obtained an accuracy of 97.6, 98.63, 100, 100% respectively using the SVM classifier for predicting late orders based on five-fold cross-validation. • This paper investigates a few DL approaches to extract features of SC data. • Both RNN and CNN are applied in the same model. • An improved CNN model has been proposed for feature extraction. • An online dataset is employed to validate the proposed DL architectures. [ABSTRACT FROM AUTHOR]

Abstract: In this paper it is introduced a recognition method for Mexican banknotes by using artificial vision. It is shown that the Mexican banknotes can be classified by extracting their color and texture features, with the RGB space and the Local Binary Patterns, respectively. We show the classification results performed with the current Mexican banknotes. We state that the proposed method can be applied to recognize banknotes of other countries which employ colors to distinguish the denominations. [Copyright &y& Elsevier]

In recent years, detection techniques using computer vision and deep learning have shown promise in assessing driver distraction. This paper proposes a fusion network that combines a Spatial-Temporal Graph Convolutional Network (ST-GCN) and a hybrid convolutional network to integrate multimodal input data for recognizing distracted driver behavior. Specifically, to address the limitations of the ST-GCN method in modeling long-distance joint interaction features and inadequate temporal feature extraction, we design the Spatially Symmetric Configuration Partitioning Graph Convolutional Network (SSCP-GCN) to model relative motion information of symmetric relationships between limbs. Specifically, we utilize densely connected blocks for processing multi-scale temporal information between consecutive frames, thereby enhancing the reuse of bottom features. Furthermore, the expression of important temporal information is augmented by the introduction of the channel attention mechanism. To tackle the problem that the Mixed Convolution (MC) combining 3D convolution with 2D convolution cannot extract higher-order timing information and has limitations in modeling global dependency relationships, we compensate for its inability to capture higher-order temporal semantic information using the Time Shift Module (TSM) without consuming additional computational resources. Additionally, the 3D Multi-Head Self-Attention mechanism (3D MHSA) is employed to integrate global spatial–temporal information of high-level features, avoiding the issue of model complexity proliferation caused by the deep stacking design of Convolutional Neural Networks (CNN). Lastly, we introduce a multistream network framework that integrates driver posture and appearance features to harness complementary advantages, enabling us to combine multimodal input features to achieve better model performance. Experimental results indicate that the accuracy of the network designed in this paper reaches 95.6% and 94.3% on ASU dataset and NTU-RGB+D dataset, respectively. The small size of the model offers the possibility for practical application of the algorithm. [ABSTRACT FROM AUTHOR]

In order to accurately predict the remaining useful life (RUL) of mechanical equipment under conditions of missing data, this paper proposes a novel integrated framework for data repair and RUL prediction, named DRTCR. DRTCR possesses the dual functionality of missing data repair and RUL prediction. Initially, considering the underlying trends and periodic fluctuations in the signals, polynomial and trigonometric functions are incorporated into the implicit feature matrix (IFM) generated by Gaussian random functions. This approach involves constructing a data repair deep learning framework primarily based on LSTM and CNN. This framework facilitates the repair of missing data. Furthermore, this paper introduces the integration of trend, correlation, and robustness within the LSTM units. A predictive unit named TCRcell is devised, which enhances the feature extraction capability of the prediction model. Additionally, a corrective unit within TCRcell is constructed, predominantly centered around discrete Fourier transform (DFT) and discrete wavelet transform (DWT) as frequency domain feature extraction techniques. This incorporation introduces the physical mechanism information into DRTCR and refines the predictive outcomes of TCRcell. The global frequency perspective of DFT aids in capturing the overall trend in signal variations, while DWT precisely captures localized temporal details in the time series, resulting in a dynamically corrected TCRcell prediction outcome that combines both global trends and local intricacies. Finally, experimental validation was conducted to assess the predictive performance and robustness of DRTCR. [ABSTRACT FROM AUTHOR]

The effective identification of bond credit risk is critical for maintaining stability in the bond market and safeguarding investor rights. This paper proposes a URP-CNN model for evaluating bond credit risk of Chinese listed companies. This model effectively combines the unique advantages of the Unthresholded Recurrence plot in extracting and representing spatial domain association information, along with the powerful performance of Convolutional Neural Networks in image feature extraction and classification tasks, thus demonstrating outstanding overall performance in tackling the complex task of predicting default of listed company bonds. The research results show that the URP-CNN model constructed in this paper improves accuracy by 1 %–10 % and F1 score by 10 %–30 % compared to other methods, ensuring accurate identification of non-defaulting enterprises while effectively reducing Type II-Error occurrence rates. Furthermore, the combination of URP method and machine learning models significantly enhances model effectiveness. Compared to single models, models combined with the URP method show varying degrees of improvement in accuracy and F1-score numerical indicators. [ABSTRACT FROM AUTHOR]

Powder bed quality is critical in Laser Powder Bed Fusion (LPBF) as defects in the powder bed will affect the quality of the forming parts. Recently, numerous powder bed defect detection methods based on semantic segmentation algorithms have been developed. However, due to the computational resource constraints and real-time requirements in industry scenarios, these heavy models face challenges in deploying into the LPBF process monitoring system. Meanwhile, existing lightweight models struggle to achieve sufficient performance on industrial data due to their limited feature extraction capabilities. To address the challenges in industrial deployment, this paper proposes a lightweight and efficient LPBF layer-wise defect detection paradigm to achieve a balance between performance and efficiency. Aiming at improving the performance of the lightweight model, this paper proposes a structured Knowledge Distillation framework, in which a pre-trained high-performance segmentation model (UNet++) is used to guide the training process of a lightweight model. To address the impact of model discrepancies on the effectiveness of Knowledge Distillation and further improve the inference speed, this paper constructs a specific lightweight model and applies Structural Re-parameterization to deeply compress the model. Moreover, we created a dataset comprising 406 images of powder-bed defects, with each image annotated at the pixel level. Massive experiments based on the dataset demonstrate the effectiveness of the proposed paradigm, with an excellent trade-off in performance and efficiency. The speed of the proposed model is approximately 5 times faster than that of high-performance models while maintaining comparable performance. • A lightweight and efficient paradigm for LPBF in-situ defect detection. • Introduced Structure Re-parameterization to enhance the Knowledge Distillation. • An excellent trade-off in performance and efficiency for model deployment. [ABSTRACT FROM AUTHOR]

• Newbie study using standard ML techniques with HAR Application and Discussions. • Activities found in Literature with the corresponding reference. • Co-occurrences between activities and sensors with the corresponding reference. • Summary and comparison among the different datasets found in Literature. • Summary of the experimentation settings with performance scores found in Literature. Human Activity Recognition (HAR) is an essential area of research related to the ability of smartphones to retrieve information through embedded sensors and recognize the activity that humans are performing. Researchers have recognized people's activities by processing the data received from the sensors with Machine Learning Models. This work is intended to be a hands-on survey with practical's tables capable of guiding the reader through the sensors used in modern smartphones and highly cited developed machine learning models that perform human activity recognition. Several papers in the literature have been studied, paying attention to the preprocessing, feature extraction, feature selection, and classification techniques of the HAR system. In addition, several summary tables illustrating HAR approaches have been provided: most popular human activities in the literature with paper references, the most popular datasets available for download (Analyzing their characteristics, such as the number of subjects involved, the activities recorded, and the sensors with online-availability), co-occurrences between activities and sensors, and a summary table showing the performance obtained by researchers. =The paper's goal is to recommend, through the discussion phase and thanks to the tables, the current state of the art on this topic. [ABSTRACT FROM AUTHOR]

The detection of underwater fish targets is critical for ecological monitoring and marine biodiversity research. However, underwater fish detection is typically constrained by problems including low image quality and variable underwater surroundings. On behalf of further improving the underwater fish detection accuracy in complex underwater environments, this paper proposes a dual-path (DP) Pyramid Vision Transformer (PVT) feature extraction network named DP-FishNet. The backbone network DP-PVT composed from the PVT Network is made up of two feature extraction paths. The first represents the Vision Transformer path, which extracts global features to enhance the distinction between the foreground and background of underwater images. The second is the convolutional neural network path, which enhances the accuracy of detecting small targets by extracting local features. Additionally, to more effectively utilize the feature information extracted by the network, this paper provides a promising solution to employ the content-aware reassembly of features (Carafe) in the feature pyramid network (FPN). The seesaw loss is utilized as a classification loss to address the problem of unbalanced samples caused by the gap in the number of fish populations. According to the experimental findings, the AP and AP50 of the DP-FishNet are 76.0% and 95.2%, respectively. In comparison to currently available advanced two-stage detection algorithms, the quantity of computation and parameters is reduced by approximately 40%. DP-FishNet strengthens the ability to extract global and local features from underwater images and enhances feature reuse. DP-FishNet can be utilized to detect fish targets in actual and complicated underwater habitats. [ABSTRACT FROM AUTHOR]

Edge detection is of critical importance for middle-level and high-level tasks in computer vision. Existing edge detection methods usually use VGG16 as the encoding network and achieve exceptional performance through transfer learning, which has the characteristics of high parameters and high computational cost. Researchers have implemented edge detection by designing decoding networks. Unlike existing methods, this paper is inspired by the effective mechanisms of edge detection in the parallel pathway of biological vision and proposes a new lightweight encoding–decoding structure for edge detection networks, which we refer to as XYW-Net. In the encoding network, we drew inspiration from the receptive field properties of X-type cells, Y-type cells, and W-type cells involved in the parallel pathway, and developed more meticulous convolutional models. Based on the structure of the parallel pathway, the three cell modules are combined into a novel encoding network and excellent feature extraction performance is obtained. Within the decoding network, this paper draws inspiration from the feature integration ability of the inferotemporal cortex (IT) and introduces a novel feature integration module to constitute the decoding network. Experiments show that the proposed network in this paper obtains the Optimal Dataset Scale (ODS) = 0.812 on the BSDS500 dataset with only 0.79 M parameters. Additionally, the model has exhibited outstanding ODS performance on various publicly available edge detection datasets, including NYUD, BIPEDv1, and Multicue. Moreover, future research could concentrate on implementing more efficient antagonistic mechanisms and devising networks grounded in higher-level visual cortical mechanisms. This could further augment the performance and efficiency of edge detection networks. We encourage researchers to delve into these directions within the field. The codes are available at https://github.com/PXinTao/XYW-Net. [ABSTRACT FROM AUTHOR]

• Pinch force strength is accurately predicted using sEMG. • Different pinch movements are accurately classified using sEMG. • Similarity of EMG features is evaluated. • An EMG feature-to-image method is proposed. The movement of intelligent EMG-driven prosthesis mainly relies on the synergy of different fingers to achieve function of grasping objects. The paper proposes a novel scheme for force prediction and movement classification about pinch between different fingers based on surface electromyography (sEMG) using machine learning. The pinch force and sEMG signals are recorded synchronously by a data acquisition device. Eight features are extracted, which are proven to have better performance in the estimation of sEMG-to-force. We present a novel feature selection method that uses the one-dimensional time series similarity assessment based on Manhattan distance to eliminate the repetitive information between features. Three optimal features carrying less repetitive information are retained. Seven machine learning algorithms are used to predict force strength. The results show that the Long Short Term Memory (LSTM) has the best performance of force prediction, achieving a R 2 of 0.9517 and RMSE of 3.2723. The paper proposes a novel method of converting EMG feature sequences to the normalized gray image in order to classify the finger movement. Five classifiers based on image feature extraction and the Convolutional Neural Network (CNN) are developed respectively. The experimental results indicate that the CNN performs best, achieving an accuracy of 97.66%. In this way, it not only realizes the accurate force prediction, but also realizes the movement classification between different fingers. The proposed methodology has the potential to realize simultaneous force and movement control of prosthetic hand. [ABSTRACT FROM AUTHOR]

Identifying the abnormal samples from a data set and determining their type are two key tasks of anomaly detection. However, the existing anomaly detection algorithms are generally faced with the defects of weak generalization ability and insufficient interpretation, the core reason of which is that they cannot mine specific features for different abnormal types. In this paper, a new anomaly detection algorithm aiming at feature selection for different abnormal types is developed. Inspired by self-supervised learning, we take the stationarity of variance changes of abnormal score similarity as a pretext task and combine it with wrapped search method. Then, the features and the corresponding parameters for different abnormal types can be screened to apply to the downstream task of anomaly integration detection. To verify the efficiency of the new algorithm, we conduct two sets of experiments to compare the new algorithm with 11 classical anomaly detection and 3 clustering anomaly detection algorithms on the data sets WDBC, WPBC and Wilt from DAMI database with the evaluation measures P@n , Adj- P@n , AP, Adj-AP and AUC. The experiment results show that, both in the identification and classification on abnormal samples, all performance measures of the new algorithm are explicitly better than that of the contrastive algorithms. Also, we apply the new algorithm to the Chinese auto insurance market, and find that the results can help managers to identify the main patterns of fraudulent claims and to summarize the feature combinations of fraud behaviors. In general, the new algorithm developed in this paper has the following advantages compared with traditional algorithms: 1) It can directly capture abnormal features and realize effective recognition of abnormal types, which effectively bridge the gap between abnormal judgement and feature screening. 2) Automatic screening of abnormal features can be completed under the condition of self-updating learning optimal strategy. 3) Only a few features are extracted from all features to reveal the abnormal characteristics, which significantly improves the interpretability and generalization ability of the algorithm and its results. In a word, through the novel self-supervised design method, feature screening is skillfully integrated into the anomaly detection task, which may provide a new way for anomaly detection research. [ABSTRACT FROM AUTHOR]

The Coronavirus Disease (COVID-19) caused a lot of mortality. The high mortality rate occurred because of the physicians' wrong or late identification of COVID-19 severity. So, developing Computer-Aided Design (CAD) systems using Artificial Intelligence (AI) techniques is critical to help physicians correctly identify the severity of COVID-19 in the early stages of the pandemic and then decrease the COVID-19 mortality percentage. In this paper, we develop a new green CAD system using a new hybrid handcrafted feature extraction algorithm and two-stage neural network architecture to grade the COVID-19 patient based on Computed Tomography (CT) scan images as having a moderate, severe, or critical infection. Because the proposed system uses handcrafted feature extraction algorithms, it consumes minimum resources and time than recent works. The proposed system consists of three phases: lesion segmentation, feature extraction, and diagnosis. Firstly, lesions from the CT scan image are manually segmented, and then three schemes are applied to extract salient features from the segmented lesions. These schemes are the Histogram of Oriented Gradients (HOG), Speeded Up Robust Features (SURF), and a new hybrid method that consists of cascading Discrete Wavelet Transform (DWT) and Gray-Level Co-Occurrence Matrix (GLCM). Then, the Cumulative Distribution Function (CDF) is computed for each scheme to extract the statistical markers. In the grading phase, a two-stage neural network approach is used. First, the extracted features are individually trained and tested for each scheme in the first neural network stage, and then the results of the first stage are combined to train and test each patient in the second neural network stage. The performance of the proposed system was assessed on a CT image dataset of 300 COVID-19-positive patients collected from the Cancer Imaging Archive (TCIA) website. The experimental results showed that our proposed system achieved 100% accuracy and kappa when the dataset was partitioned into 80% for training and 20% for testing. Also, it achieved 95.67% ± 0.47, 99.33% ± 0.77, and 100% ± 0 accuracies and 93.48% ± 0.74, 98.997% ± 1.16, and 100% ± 0 kappa values when the data was organized using 2, 4, and 10 folds, respectively. A green complexity algorithm analysis shows that this proposed system takes O(n) time complexity and 1 h and 20 min to train and test all cases. The performed green complexity analysis shows that the proposed system consumes 117.80 g Carbon Dioxide Equivalent (CO2e), 130.80 Wh, and 0.13 tree months for the carbon footprint, the energy needed, and the carbon sequestration, respectively. These results show that the proposed work consumes fewer resources and provides a green CAD system. • This paper proposes a CAD system for COVID-19 severity prediction. • The proposed system uses a two-stage neural network architecture. • The proposed system consists of two phases: feature extraction and severity identification. • A green complexity analysis is performed to prove the greenness of the proposed system. [ABSTRACT FROM AUTHOR]

Epilepsy is one of the most paramount neurological diseases, affecting about 1% of the world's population. Seizure detection and classification are difficult tasks and are ongoing challenges in biomedical signal processing to enhance medical diagnosis. This paper presents and highlights the unique frequency and amplitude information found within multiple seizure types, including their morphologies, to aid the development of future seizure classification algorithms. Whilst many published works in the literature have reported on seizure detection using electroencephalogram (EEG), there has yet to be an exhaustive review detailing multi-seizure type classification using EEG. Therefore, this paper also includes a detailed review of multi-seizure type classification performance based on the Temple University Hospital Seizure Corpus (TUSZ) dataset for focal and generalised classification, and multi-seizure type classification. Deep learning techniques have a higher overall average performance for focal and generalised classification compared to machine learning techniques, whereas hybrid deep learning approaches have the highest overall average performance for multi-seizure type classification. Finally, this paper also highlights the limitations of the TUSZ dataset and suggests some future work, including the curation of a standardised training and testing dataset from the TUSZ that would allow a proper comparison of classification methods and spur advancement in the field. [ABSTRACT FROM AUTHOR]

Automatic fault diagnosis in many systems is performed via the analysis of vibration data in the time or frequency domain. In the literature, many approaches for extracting featured from signals for fault diagnosis have been proposed. In this paper, we apply a variety of transform functions including Fourier, Wavelet, etc. to extract features from the vibration data, which includes the statistical features and some features automatically extracted via Convolutional Neural Networks (CNNs). For each of these feature extraction approaches, a learning algorithm is trained to diagnose the faults and their results are aggregated in an ensemble machine learning algorithm. The weights of the base learner algorithms are optimized via an evolutionary algorithm to achieve the best-weighted voting scheme. The architecture of CNNs has a significant effect on the performance of the algorithm, thus, in this paper, an evolutionary algorithm is proposed to find the best architecture for CNNs in fault diagnosis. The CNNs are trained via gradient descent algorithms which suffer from getting stuck in local optima. To manage this, we propose an evolutionary algorithm that benefits from the speed of gradient descent and the global search of evolutionary algorithms. The proposed algorithm is tested on a number of benchmark problems and the experimental results are presented. • An ensemble algorithm is proposed for fault diagnosis. • Different feature extractors are exploited. • An evolutionary training algorithm is proposed. [ABSTRACT FROM AUTHOR]

Transfer learning (TL) for bearing fault diagnosis has achieved remarkable success under different operating conditions, even with a limited amount of data. However, it is hard to implement TL from high-speed to very-low-speed (VLS) bearings due to the large domain discrepancies. Two main factors can bring about domain discrepancies between high-speed to VLS bearings, specifically, 1) revolution discrepancy and 2) peak discrepancy that arise from different operating speeds and different bearing specifications, respectively. These factors disturb TL's feature extraction process as it seeks speed-invariant and fault-related characteristics. Thus, this paper newly defines two metrics – revolution discrepancy D R and peak discrepancy D P – to quantify the domain discrepancies between different bearing systems, including consideration of rotational speed and specification differences. Furthermore, domain alignment techniques are proposed to reduce these domain discrepancies, thereby enabling adequate TL for bearing fault diagnosis. In this research, the revolution matching module (RMM) is developed to be helpful for extracting speed-invariant features by matching the revolution information under different rotational speeds. Further, the peak matching module (PMM) is developed to extract fault-related features by matching the defect peak information from different specifications. The proposed method is validated using experimental bearing datasets under various speeds. The results demonstrate that the proposed metrics can quantify the discrepancies between different bearing systems, and that the use of RMM and PMM can successfully improve TL performances for VLS bearings even under insufficient data conditions. [ABSTRACT FROM AUTHOR]

Road damages, such as potholes and rutting, must be addressed in their early stages to prevent accidents and minimize maintenance costs. The presence of these damages not only causes discomfort to passengers but also accelerates vehicle deterioration. Improper material mixtures and inadequate maintenance practices often contribute to road damage. In recent years, numerous machine learning (ML) algorithms have been developed with a predominant focus on binary classification, specifically targeting potholes. However, roads are susceptible to various damages beyond potholes, necessitating a comprehensive approach for effective solutions. This study introduces a novel algorithm utilizing a random forest (RF) classifier for multi-class classification of road damages. The proposed algorithm is validated through rigorous simulation and field studies. Unlike previous models, this approach embraces a comprehensive perspective by considering a wider range of damages, thereby facilitating a more nuanced and inclusive process for the classification of road damages. In the field study, data has been collected using a set of uni-axial accelerometers and a smartphone camera mounted on a test vehicle. The data has been processed using the sliding window technique, and features have been extracted from each window to train the RF classifier. In this process, an optimal window size has been determined and employed to enhance the effectiveness of feature extraction for training the RF classifier. The proposed algorithm demonstrates significant accuracy in both simulation and field studies, with notable performance in identifying and classifying road damages. The algorithm's outcomes are utilized to estimate a detailed cost for repairing the identified damages. This paper showcases proficiency in addressing various damages, offering valuable insights for the implementation of cost-effective road maintenance strategies. • A novel RF-based algorithm is employed for the multi-class classification of road damages. • Validated through simulation and field studies, the algorithm achieves notable performances. • Compared the evaluation metrics with other existing classifiers. • Features required for classification are extracted using the sliding window technique. [ABSTRACT FROM AUTHOR]

Common subspace learning methods only utilize local or global structure in feature extraction, and cannot obtain the global optimal discriminative projection matrix. For this reason, this paper proposes a discriminative sparse subspace learning method based on the manifold regularization framework (DSSL-MR), which introduces the graph Laplacian matrix that reflects the intrinsic geometric structure of the sample as a penalty term. DSSL-MR simultaneously uses both sub-manifold and multi-manifold information of samples for obtaining optimal projection to enhance the discriminability of different classes in subspace. DSSL-MR uses the sparse property of the L 2 , 1 -norm to constrain the projection matrix, which can eliminate redundant features and select features that are significant for classification. It is a linear supervised method, which belongs to the Fisher discriminant analysis framework. Experimental results on multiple real-world datasets show that the algorithm is very effective in classification and has high recognition rates. • Subspace learning and spectral graphs are combined into a unified learning framework. • The learned projection maps samples to the subspace with optimal class separation. • The manifold regularization framework fusing decision boundary and subspace learning. • The sample structure is analyzed by fusing multimanifold and submanifold information. [ABSTRACT FROM AUTHOR]

It is a crucial means for humans to perceive geomorphic features and landscape architectures by classifying ground objects in hyperspectral images (HSIs). Currently, the exponential development of neural networks has provided a powerful support for the accurate HSI classification. However, existing neural network-based methods usually rely solely on the data to drive the classification model, lacking attention to valuable land-cover distribution knowledge in HSIs. In view of this, to utilize hyperspectral data and distribution knowledge simultaneously, a data and knowledge-driven deep multiview fusion network based on diffusion model (DKDMN) is proposed in this paper. DKDMN extracts knowledge from unlabeled data in HSIs through a diffusion model-based knowledge learning framework (DMKLF), and combines raw hyperspectral data with the acquired knowledge through a designed deep multiview network architecture (DMNA) to mine complicated land-cover distribution information and reflect sample relationships. First, the proposed DMKLF utilizes the data distribution reconstructed by the diffusion model as a knowledge source for one view to enhance the network cross-sample awareness ability. On the other hand, the original HSI patches are considered a data source for another view, which co-drives DMNA with the unsupervised diffusion knowledge extracted by DMKLF to perform effective feature extraction. Second, taking into account the characteristics of each view and the feature similarity between these two views, a joint loss function specifically for DMNA is suggested to minimize the difference between the model predictions and the real labels. Finally, a multi-backbone integration classification framework (MBICF) is designed by deeply fusing three vision architectures to capture multi-scale spectral features and local–global features, thereby achieving pixel-wise classification effectively. Experimental results on four publicly available HSI datasets demonstrate that the proposed DKDMN achieves competitive classification accuracy compared with other state-of-the-art methods. For instance, the proposed DKDMN achieves an overall accuracy improvement of 1.62% and 2.18% on the Indian Pines and Salinas Valley datasets, respectively, compared to the multiple vision architecture-based hybrid network (MVAHN). The related code will be released at https://github.com/ZJier/DKDMN. • Designing a data and knowledge-driven HSI classification method. • Learning unsupervised spectral spatial distribution knowledge. • Combining data source and knowledge source by deep multiview network architecture. • Realizing HSI classification by multi-backbone integration classification framework. [ABSTRACT FROM AUTHOR]

The task of multimodal image fusion aims to preserve the respective advantages of each modality, such as the detailed texture information from visible light images and the salient target information from infrared images. However, images in real environments are often not perfect high-quality images but are affected by various degradation factors such as noise, blur, and glare. Existing unsupervised multimodal image fusion algorithms use undegraded source images as input to fusion networks and generate high-quality fusion images through complex fusion strategies. This single linear learning approach from high-quality to high-quality cannot be applied to the fusion task of degraded multimodal images in real environments. To address this issue, this paper proposes an end-to-end multimodal image fusion network based on a high-order local degradation model (MFHOD). Firstly, inspired by the idea of probabilistic degradation, we propose a high-order local random degradation model (HODM), which inputs the source multimodal images into the degradation model to obtain degraded images before feeding them into the network. Secondly, we design a simple and efficient dual-branch feature extraction encoder to extract deep features from images. Then, from the perspectives of image pixels, brightness, and gradients, we propose an improved composite loss composed of multiple loss functions to constrain network training. Finally, we propose a L 2 -norm fusion strategy for preserving brightness information in low-light nighttime images. Our MFHOD demonstrates good performance on infrared and visible light image datasets as well as medical image datasets. Experimental results show that MFHOD can effectively suppress the effects of glare, noise, and smoke in adverse environments, and also improve the quality of fusion images in low-light and nighttime environments. • Linear fusion methods cannot fully train the feature extraction of neural networks. • Two-stage high-order degradation models significantly boost network nonlinearity. • End-to-end network simplifies multimodal image fusion process. • MFHOD boosts brightness and contrast, suppressing glare, noise, and smoke effects. [ABSTRACT FROM AUTHOR]

The open-source nature of Android, along with its coarse-grained permission management and widespread use, has heightened its vulnerability to malware threats. However, many traditional approaches face limitations in extracting comprehensive features, which hinders accurate and efficient Android malware detection. In this paper, we propose an effective hybrid approach combining an improved multi-scale convolutional neural network (MSCNN) with residual networks (ResNet) to defend against Android malware. The approach comprises an enhanced feature extraction network and a detection network. Initially, we introduce MSCNN, a novel deep learning model for comprehensive feature extraction. The MSCNN extends into three branches at distinct levels, concatenating features from low to high dimensions. This multilevel structure preserves rich semantic features while avoiding complex feature selection and analysis. Further, ResNet is employed as the detection network, and the hybrid models' performance is evaluated by comparison with a single ResNet. Finally, we validate the effectiveness of our approach by comparing our experimental results with state-of-the-arts. The experimental results show our approach effectively detects Android malware with high accuracy (99.20%) and precision (99.49%), and utilizing MSCNN as a multilevel feature extraction network significantly enhances the performance of the hybrid models, particularly, the F1-Score of MSCNN+ResNet18 increases by 4.8%. • Propose an improved MSCNN model extracting features from various data levels. • Design hybrid models combining MSCNN with GRU, ResNet18, ResNet34 and ResNet50. • MSCNN acts as upper feature extraction layer. • GRU, ResNet18, ResNet34 and ResNet50 as detection network, respectively. • Our approach detects Android malware, and enhances the hybrid model performance. [ABSTRACT FROM AUTHOR]

• A transform-based feature reconstruction network is designed for missing modalities. • The model reconstructs missing element features to improve the robustness. • Modality-invariant representation is used in missing modalities. • Modality-invariant temporal representation is used to mitigate heterogeneity. • Experimental results show the superior performance of our proposed method. The investigation of missing modalities aims to extract valuable feature information from missing multi-modal data, it is a focal point in multi-modal learning. Existing missing modalities processing methods primarily focused on multi-modal fusion schemes to achieve optimal performance, but they faced the following two key challenges: (1) how to improve the robustness of incomplete multi-modal sequence representation, (2) how to effectively learn modality-invariant representations to mitigate heterogeneity between modalities. In this paper, we propose a m odality- i nvariant t emporal representation learning-based f eature r econstruction net work called MIT-FRNet for the missing modalities to tackle these challenges. In the MIT-FRNet, we first extract the latent features for each modality considering intra-modality and inter-modality, and then introduce an encoder-decoder framework to address the first challenge, it reconstructs missing element features via taking the incomplete modal sequences as input and implementing the inter-modal and cross-modal attention mechanisms for feature extraction. And then, through treating each timestamp as a single Gaussian distribution, we design a fine-grained similarity constraint based on distribution-level modality-invariant representations to learn effective modality-invariant representations, thereby addressing the second challenge. Finally, the efficiency of proposed model is validated through the classification results after multi-modal fusion that involves using a gate encoder to pass it and followed by a vector fusion to fuse it. Extensive experiments on public benchmark datasets demonstrate that the proposed MIT-FRNet method achieves promising results under varying missing rates while exhibiting good convergence. [ABSTRACT FROM AUTHOR]

The goal of pansharpening methods is to complement the spectral and spatial information contained in Multi-spectral (MS) and panchromatic (PAN) images to obtain the desired High-resolution multispectral (HRMS) image. The existing majority of pansharpening methods either extract feature information separately from the MS image and PAN image, or extract feature information after concatenating the MS image and PAN image. However, the entire extraction process lacks the utilization of complementary information and tends to generate redundant information, thereby leading to the loss of certain important information during the extraction process, which in turn affects the overall performance. In order to better utilize the complementary information between the MS image and PAN image and enhance the interpretability of the network, we propose the Deep Unfolding Convolutional-Dictionary Network (DUCD) for pansharpening in this paper. This network fully integrates complementary information between the MS image and PAN image to generate the final fused image. The entire network structure consists of two parts: The encoder and the decoder. In the encoder part of the network, we clarify the common and unique feature information between MS and PAN images by constructing an observation model. Simultaneously, we use the approximate gradient algorithm to continuously optimize the model and iteratively unfold it into a deep network structure. In the decoder part of the network, we concatenate the obtained common and specific information from MS and PAN images and pass them through convolutional and activation layers. Subsequently, they are input into the introduced Frequency Domain-based Transformer (FDT) module and an information-lossless inversible neural network(INN). This provides a more efficient method for establishing long-range dependency relationships between feature extraction and feature fusion. To demonstrate the effectiveness of our proposed method, we conduct extensive experiments on three benchmark datasets QB, GF2 and WV3. Experimental results show that our method outperforms the current SOTA Pansharpening methods in terms of performance. • Convolutional-Dictionary Network, Pansharpening. • Prior knowledge, Proximal gradient algorithm, strong interpretability. • Inversible neural network, Attention mechanism. [ABSTRACT FROM AUTHOR]

The proliferation of depth cameras and LiDAR sensors in actual industrial environments has fueled the pursuit of an effective and efficient 3D point cloud model that enables us to perceive and interact with the physical world. However, the intrinsic complexity of 3D semantic information poses significant challenges to model design, including spatial rotation invariance and irregular point cloud structure, which fundamentally impact the representation and behavior of 3D point cloud systems. Existing have either heavily relied on labeling information in a supervised learning setting or failed to effectively capture the inherent patterns of the 3D point clouds within a self-supervised learning framework, leading to poor performance in specific downstream tasks. To address these limitations, this paper introduces a self-supervised framework, D ual- C ross C ontrastive Neural N etwork (DCCN) for 3D point cloud self-supervised representation learning. DCCN leverages cross-view, cross-network, and domain-specific knowledge distillation to enhance the extraction of hidden features from point clouds and fully exploit the capabilities of the encoder. Our DCCN employs a pseudo-Siamese network consisting of an online network and a target network, facilitating knowledge interaction and distillation. The method extracts internal states from augmented 3D point cloud by learning cross-view relationships and optimizes model parameters through intra-modal cross-network learning. We incorporate a momentum-updating mechanism without shared weights in the Siamese network architecture to distill knowledge and enhance the role differentiation the online and target networks. Experimental results demonstrate that our approach outperforms a range of supervised and self-supervised learning methods across a series of downstream tasks consisting of four tasks in three representative datasets. Ablation studies validate the component-wise effectiveness of cross-view, cross-network, and moment-updating learning objectives in achieving superior point cloud representation. The overall findings establish our method, DCCN, as an effective solution for 3D point cloud representation learning in real-world applications. [ABSTRACT FROM AUTHOR]

The application of virtual reality (VR) technology in rehabilitation is a research area that has been attracting remarkable attention recently, especially as an intervention with learning and developmental disorders such as dyslexia and autism. Multi-modal VR-based rehabilitative systems can provide a rich experience to patients by responding to various types of bio-signals. This paper proposes a framework for incorporating voice input in rehabilitative games. The proposed framework equips collaborative learning environments with voice based interaction which allows for a more dynamic flow of therapy sessions between physically impaired patients and specialists in a rehabilitative VR environment. The proposed framework's pipeline begins with a speech recognition module to parse utterances into sentences. Next, comes the feature extraction module to identify the intent of the user and entities as well. Finally, we modulate the 3D environment in Unity. The framework was designed in a way that guarantees smooth integration with any game or 3D environment. Several technologies were tested in terms of accuracy and speed to increase the overall performance of the pipeline. To the best of our knowledge, this research is the first to do the following: 1) build a dataset of instructions, inspired by a product already used in the healthcare market, 2) based on which we could realize a multi-modal system, with voice as one of the modalities, for the context of VR-based rehabilitation. We evaluate every stage in the system pipeline, and we show that we can achieve considerable results on each of these stages including the speech recognition module, with a 5.64% word error rate, and on intent classification where the BiLSTM model outperforms BERT in the Named Entity Recognition by 11% in terms of the F1-score. We envisage the proposed system to lend itself well to emerging healthcare frameworks in the Metaverse. [ABSTRACT FROM AUTHOR]

Highlights • A novel modeling strategy for Decision Trees is proposed. • Interactive basis functions induce oblique and non-linear decision boundaries. • Benchmarking shows potential gains in accuracy and/or parsimony of models. • Empirical examples illustrate the application of the modeling strategy. • Decision charts are introduced to improve the interpretability of models. Abstract Decision Trees (DTs) are a machine learning technique widely used for regression and classification purposes. Conventionally, the decision boundaries of Decision Trees are orthogonal to the features under consideration. A well-known limitation of this is that the algorithm may fail to find optimal partitions, or in some cases any partitions at all, depending on the underlying distribution of the data. To remedy this limitation, several modifications have been proposed that allow for oblique decision boundaries. The objective of this paper is to propose a new strategy for generating flexible decision boundaries by means of interactive basis functions (IBFs). We show how oblique decision boundaries can be obtained as a particular case of IBFs, and in addition how non-linear decision boundaries can be induced. One attractive aspect of the strategy proposed in this paper is that training Decision Trees with IBFs does not require custom software, since the functions can be precalculated for use in any existing implementation of the algorithm. Since the underlying mechanisms remain unchanged there is no substantial computational overhead compared to conventional trees. Furthermore, this also means that IBFs can be used in any extensions of the Decision Tree algorithm, such as evolutionary trees, boosting, and bagging. We conduct a benchmarking exercise to understand under which conditions the use of IBFs can improve model the performance. In addition, we present three empirical applications that illustrate the approach in classification and regression. As part of discussing the empirical applications, we introduce a device called decision charts to facilitate the interpretation of DTs with IBFs. Finally, we conclude the paper by outlining some directions for future research. [ABSTRACT FROM AUTHOR]

Highlights • Imposing sparsity constraint on the activations along the depth of feature maps. • Feature map dropout in the output of the last convolutional layer. • The ensemble of multiple models obtained from the periodic learning rates. Abstract Deep convolutional models have been able to get extraordinary results in visual, speech and textual processing domains. Nevertheless, the complexity of manifold of images in data space necessitates the deep networks to have a large number of parameters and subsequently these deep models prone to redundancy and overfitting. One of the most effective methods to tackle this problem is devising special regularization methods in the context of convolutional models. In this paper, after presenting and discussing the most important models in the field, sparse feature maps are proposed and employed as a penalty term in the cost function. This conducts the learning process to construct kernels so that the feature maps at every point are sparse along the depth. A sparse representation is able to adapt to the varying level of information of natural images and can help to extract more independent and informative representations. Also, the DropMaps concept is proposed and employed in the last convolutional layer of the model. This technique applies dropout on the feature maps that causes coincidence of feature maps to be avoided. As is shown in this paper, sparse feature maps and DropMaps can handle the problem of overfitting in large models for tiny images. We have studied the effect of the sparsity rate on the accuracy of the model, and it is observed the accuracy of the test dataset reaches its maximum at a sparsity rate of 0.05. Moreover, by designing appropriate learning rate curves, we were able to obtain ensemble machines with much less cost for training. It is noticeable that the test accuracy is higher than the validation accuracy of the ensemble, indicating that the model has not overfitted. In the input of the model, a random online preprocessing layer is employed for the training phase that helps regularization of the model. Comparing input space and feature space of the model we found that the proposed network is able to successfully separate images of different classes. Finally, testing the proposed model with MNIST dataset has shown that the test set can be classified with accuracy 99.75. The same test with CIFAR 10 dataset attained an accuracy of 94.05. [ABSTRACT FROM AUTHOR]

Highlights • We propose a representation based on Fuzzy Temporal Windows for binary-sensors. • Long Short-Term Memory is deployed as a means of sequence classifier. • A balanced training is included to build an ensemble of activity-based classifiers. • The proposed approach is evaluated and benchmarked against previous approaches. Abstract There are approaches that successfully recognize activities of daily living by using a trained classifier on feature vectors created from binary sensor data. Although these approaches have been successful, there are still open issues such as the evaluation of multiple temporal windows, ensembles of classifiers or unbalanced classes which need to be addressed in order to improve the performance of the real-time activity recognition process. In this paper, we present a methodology for Real-Time Activity Recognition based on the diverse fields of Machine Learning, including Fuzzy Logic and Recurrent Neural Networks. The methodology uses a long-term and short-term representation of binary-sensor activations based on Fuzzy Temporal Windows. The paper proposes an ensemble of activity-based classifiers for the purposes of balanced training, where each classifier in the ensemble is a Long Short-Term Memory. The approach was evaluated using two binary-sensor datasets of daily living activities and benchmarked against previous approaches based on the combination of sensor activation features. [ABSTRACT FROM AUTHOR]

Highlights • Extracting features using syntactic rules from pros/cons reviews. • Determining the opinion's polarity according to their emotion strength. • Comparing results with state-of-the-art approaches in different tasks. Abstract Sentiment knowledge extraction is a growing area of research in the literature. It helps in analyzing users’ opinions about different entities or events, which can then be utilized by analysts for various purposes. Particularly, feature-based sentiment analysis is one of the challenging research areas that analyzes users’ opinions on various features of a product or service. Of the three formats for the product reviews, our focus in this paper is limited to analyzing the pros/cons type. Due to the nature of pros/cons reviews, they are mostly concise and follow a different structure from other review types. Therefore, specialized techniques are needed to analyze these reviews and extract the customers’ discussed product features along with their personal attitudes. In this paper, we propose the Pros/Cons Sentiment Analyzer (PCSA) framework that exploits dependency relations in extracting sentiment knowledge from pros/cons reviews. We also utilize two different lexicons to ascertain the polarity strength of the extracted features based on the customers’ opinions. Several experiments are conducted to evaluate the performance of PCSA in its different phases. [ABSTRACT FROM AUTHOR]

Emotion is an intense mental experience often manifested by rapid heartbeat, breathing, sweating, and facial expressions. Emotion recognition from these physiological signals is a challenging problem with interesting applications such as developing wearable assistive devices and smart human-computer interfaces. This paper presents an automated method for emotion classification in children using electrodermal activity (EDA) signals. The time-frequency analysis of the acquired raw EDAs provides a feature space based on which different emotions can be recognized. To this end, the complex Morlet (C-Morlet) wavelet function is applied on the recorded EDA signals. The dataset used in this paper includes a set of multimodal recordings of social and communicative behavior as well as EDA recordings of 100 children younger than 30 months old. The dataset is annotated by two experts to extract the time sequence corresponding to three main emotions including “Joy”, “Boredom”, and “Acceptance”. The annotation process is performed considering the synchronicity between the children's facial expressions and the EDA time sequences. Various experiments are conducted on the annotated EDA signals to classify emotions using a support vector machine (SVM) classifier. The quantitative results show that the emotion classification performance remarkably improves compared to other methods when the proposed wavelet-based features are used. [ABSTRACT FROM AUTHOR]

There are lots of research papers for ASC, and in recent years it is rapidly increasing. DCASE also provides different types of competition for the submission of several papers to solve the various tasks of ASC and it is the opportunity for the research scholars either to participate in those competitions or to provide the enhanced model for ASC. This paper provides details about the various recent approaches along with the block diagram used for pre-processing required before model development for ASC. It also includes a description of different recent techniques used for the classification of different sounds for ASC tasks. The comparative analysis for different recent available techniques both for pre-processing and classification has been done and summarized in this paper. It also describes the contributions towards the survey on ASC by comparing this paper with some existing survey papers based on several parameters like functionality described separately, results from description with quantifiable value, a dataset with proper quantifiable analysis, and pictorial representation of model discussed, etc. Finally, considering the benefits for eminent research scholars, this paper has also focused on the details for future directions for both pre-processing and classification for ASC. [Display omitted] • A systematic review on pre-processing and classification techniques for acoustic scenes. • Comparative analysis of pre-processing and classification techniques. • Discussion on different processing techniques for classification. • Identifies the future directions for further research in the area. [ABSTRACT FROM AUTHOR]

• The feature representation of the network was extracted from a randomly searched sequence of vertices. • In networks of scientific documents, representation learning needs to consider similarity and citation relationships. • This paper proposed a method to search and learn multiple paths connecting similar documents in a scientific document network. • From scientific papers and patent datasets, we found empirical evidence that our algorithm is suitable for representation learning. • This paper can contribute to preparing the basis for inducing innovation in science and technology from scientific documents. A network is a structure that can represent an organic relationship of observations. Network representation learning has the advantage of extracting latent features in a network. In recent years, various algorithms have been developed to embed network representation in a low-dimensional feature space. Conventional algorithms learn a sequence of vertices that are randomly searched from a network and convert a vertex into a vector. This study focused on the representation learning of document networks. Thus, this study proposes a scalable network representation learning algorithm that preserves the structure and content of the document network and reduces data sparsity. To do this, we used the shortest walks to extract a sequence of vertices that reflected semantic similarity from a network. The shortest walk is a method for finding several shortest paths that connect similar vertices. After this, the sequence of vertices was inputted into the language model, assuming that the sequence was a sentence. Scientific papers and patent datasets provided empirical evidence for the validity of our approach. The former, the bibliography network of scientific paper, consists of 17,716 nodes and 105,734 edges. The latter is a patent describing a technology for energy, representing a network of 24,894 nodes and 28,314 edges. The results showed that the proposed algorithm exhibited the lowest error of the tested methods in multi-class classification and link prediction tasks. The results of these tasks provided evidence that our algorithm is suitable for extracting feature representations of document networks. [ABSTRACT FROM AUTHOR]

Feature extraction networks play a crucial role in classification algorithms. However, most convolutional neural networks use the form of residual block stacking to extract downstream features of images, and this simple form is not sufficient to obtain better classification performance. Therefore, this paper proposes a YOLOX detection framework based on an improved ConvNeXt which incorporates many advantages of Swin-Transformer with efficient classification performance. To address the problems of unstable IOU convergence and inaccurate regression coordinates of the algorithm, this paper adopts Manhattan distance as the new penalty term, and the new loss function combines the advantages of Euclidean distance and Manhattan distance, which has no gradient explosion in the early training period and can drop smoothly to a stable point in the late training period. The proposed algorithm is robust to negative effects such as dust, light, shadows, and occlusions, and is well-suited for masked-face detection in industrial and public scenarios. The proposed model is tested on the VOC dataset and the AIZOO open-source masked-face dataset, obtaining 5.58% AP improvement on the VOC dataset and reaching 96.665% AP on the masked-face dataset, which is better than other algorithms. • Designing a deep neural network to use for masked-face detection. • Anchor-free model is used to avoid time-consuming post-processing. • The improved ConvNeXt network improves classification performance. • Adopting Manhattan distance-based localization loss makes training stable. • The Mosaic data augmentation strategy alleviates the overfitting problem. [ABSTRACT FROM AUTHOR]

As a common state feature extraction method, spectral analysis methods have been widely used in various fields. However, due to the influence of fixed envelope, the existing spectral analysis method is not outstanding in extracting state features, and it is difficult to extract state features under strong interference. In this paper, a new feature extraction method, called global optimal Ramanujan spectrum (GORS), is proposed. GORS method innovatively uses iterative generalized envelope to improve the accurate enhancement of weak fault features and avoid the defect of unstable feature extraction performance caused by fixed envelope. Meanwhile, an evaluation index without pseudo-monotonicity, logarithmic Ramanujan spectral signal-to-noise ratio (lg(RSSNR)), is defined for the first time, which is used to help select the optimal envelope results and adaptively determine the optimal iterative envelope number, so as to realize the global optimization of spectral analysis results and maximize the ability of state feature extraction. (Pseudo-monotonicity of feature index (PMFI) is defined as: there is no stable correspondence between the change law of characteristic index and the information content of state characteristics). The evaluation results of fault characteristic strength (FCS) by different methods show that GORS method has good anti-interference ability and is an effective state feature extraction method. [ABSTRACT FROM AUTHOR]

Spatial–temporal feature extraction methods have been widely used for gesture classification using surface electromyography (sEMG) signals. Nevertheless, prevalent methodologies in this realm are constrained by limitations stemming from their simplistic single-branch architectures, which impose sequential constraints and facilitate unidirectional information flow during the feature extraction phase. To enhance the accuracy of gesture recognition by comprehensively capturing intricate spatial–temporal features in sEMG signals, this paper introduces SGRN, a novel multi-branch spatial–temporal feature extraction network. SGRN integrates a meticulously crafted Spatial Feature Extraction Network (SNet), Temporal Feature Extraction Network (TNet), and Spatial–Temporal Feature Fusion Network (STNet), enabling comprehensive multi-dimensional feature extraction and seamless multi-branch information fusion. Distinct from prior spatial–temporal fusion methods, SGRN concurrently performs spatial feature extraction and temporal modeling, followed by multi-branch fusion, thereby harnessing the full potential of the multi-branch architecture to boost model performance. Extensive experiments on four datasets conclusively demonstrate SGRN's efficacy and superiority over state-of-the-art models, presenting a promising avenue for prosthetic control and muscle-computer interaction. • Introducing a surface electromyography-based gesture recognition network. • The network enhances gesture recognition by fusing spatial and temporal features. • Evaluations show the proposed network's superior accuracy and generalization. [ABSTRACT FROM AUTHOR]

Dog barks are a crucial means for dogs to express their emotions, conveying needs and emotional states such as anger, sadness, or excitement. With the increasing number of pet dogs, understanding their emotional states is essential for better care and interaction. However, accurately interpreting the emotional content of dog barks is often challenging for humans. Therefore, developing a technology for recognizing the emotions in dog barks is of great importance. In this paper, we establish a dataset for dog bark emotion recognition and propose a method using the Synchrosqueezing Short-Time Fourier Transform (SST_STFT) to extract the time–frequency features of dog barks. Considering the characteristics of dog barks, we design an emotion recognition model based on Mamba. This model leverages the state-space model (Select SMM) for global modeling of long sequence features, enabling rapid and effective processing of time–frequency features and accurate perception of their variations. Experiments on the self-built DogEmotionSound dataset, as well as the public IEMOCAP and UrbanSound8K datasets, demonstrate that our model outperforms existing state-of-the-art sound recognition models across various evaluation metrics. The recognition accuracies on the three datasets are 91.97%, 95.36%, and 67.25%, respectively. The code is open-source at https://github.com/yangcjya/A-barking-emotion-recognition-method-based-on-Mamba.git. • We built the DogEmotionSound dataset by ourselves. • We proposed a dog barking emotion recognition model based on Mamba. • We proposed a high and low frequency feature extraction module (HL-FE). • We proposed to use the SST_STFT method to extract the feature of sound. [ABSTRACT FROM AUTHOR]

Relation extraction is significant for text structuring of colorectal cancer (CRC) pathological reports to facilitate doctors' disease diagnoses. Although many relation extraction methods have been extensively studied for various natural language processing applications, they cannot be well transferred to be applied for CRC pathological reports since CRC pathological reports have some unique characteristics. To this end, a deep learning framework is designed in this paper to extract entity relations in CRC pathological reports, which is based on an encoder–decoder architecture with entity-aware feature orthogonal decomposition. Specifically, to effectively extract semantic features of long and short entities, a two-stream encoder is designed based on an edge-aware convolutional neural network and a dimension-aware dilated convolution residual network. To alleviate the influence of the blending of subject–object features, entity-aware feature orthogonal decomposition is designed to decompose the extracted semantic features into three types, i.e. subject features, object features and subject–object shared features. A stage-wise cross entropy loss is proposed to well train the network. Comparison experiments indicated that our designed network performs well on CRC pathological texts with the performance of 92.3% F1 score, 93.1% Precision, and 91.5% Recall, outperforming the existing relation extraction models. [Display omitted] • Joint relation extraction for CRC pathological reports via deep learning. • Edge-aware CNNs and dimension-aware DCRN for extracting entity features. • Entity-aware feature orthogonal decomposition for decomposing entity features. • Stage-wise cross-entropy loss is proposed to well ensure the network training. • Perform well on real CRC pathological reports with the 92.3% F1 score. [ABSTRACT FROM AUTHOR]

Accurate segmentation of honeycomb lung lesions from lung CT images is crucial for the clinical diagnosis of a wide range of lung diseases. Existing semi-supervised segmentation methods ignore the rich texture feature information in honeycomb lung CT images, resulting in inaccurate predictions in the lesion region. Therefore, this paper proposes a semi-supervised honeycomb lung lesion segmentation framework using texture feature-aware consistency. The framework consists of a shared encoder, a main decoder, and three auxiliary decoders. The different outputs produced by the decoders are used to compute the Grey Level Co-occurrence Matrix and extract texture features, respectively, and further to compute the loss of texture feature-aware consistency. In pseudo-labeling supervision, the uncertainty of each output is computed to bootstrap the pseudo-labeling supervision loss. For the small lesion recognition omission problem, a pseudo-label refinement module is constructed to reuse the multi-scale features of the encoder, focusing on small-scale lesions, and refining the pseudo-labels. Experimental results show that the proposed method outperforms other methods in the honeycomb lung segmentation task, achieving a Dice coefficient of 77.62 using 20% labeled data. Additionally, experiments on publicly available datasets further demonstrate the generalizability of the proposed method. The code is available at https://github.com/Oran9er/TFC-Net-main. [ABSTRACT FROM AUTHOR]

In the context of smart manufacturing, modern industrial processes are becoming increasingly complex in terms of process flows, product varieties, and performance indicators (PIs). Performance-driven process monitoring attracts extensive attentions in recent years. However, most methods require spatio-temporal correspondence between process variables and PIs, and rarely consider the correlation among various PIs. In this paper, a spatio-temporal feature extraction network-based multi-performance indicators synergetic monitoring framework is presented. Firstly, considering the missing data in PI measurements, a weighted sum of tensor nuclear norm (WSTNN) based batch-process data completion approach is developed, which can adeptly handle local missing and incomplete data issues and establish the spatio-temporal correspondence for subsequent modeling. Secondly, for a specific PI, a new canonical variate analysis embedded spatio-temporal convolutional network (CVA-STCN) is designed to extract the PI-related features with spatio-temporal dependency. Thirdly, considering the dynamic interaction of multiple PIs, a third-order feature tensor is established to perform the future fusion, and the correlations among various PI-related features are explored via tensor decomposition. Finally, a hierarchical multi-performance indicators synergetic monitoring model is developed over several subspaces. The proposed method is verified on Tennessee Eastman process and an actual hot strip mill process. Overall, the monitoring performance of the proposed method outperforms the traditional ones, in terms of higher fault detection rates and lower false alarm rates. Moreover, the information provided by the multi-subspace synergetic monitoring charts can offer valuable guidance to field engineers. [ABSTRACT FROM AUTHOR]

Image splicing can easily be used in illegal activities, which may have negative impact on society. Therefore, detecting splicing images and precisely localizing tampered regions are challenging tasks. In this work, we propose a dual-branch Multi-Scale Noise-Guided Progressive Network (MSNP-Net). The multi-resolution branch extracts deep semantic features of images while suppressing redundant noise. A multi-scale noise-guided branch is designed to capture more subtle tampering artifacts and guide the network to strengthen the learning of spatial structure features. The two branches complement and restrict each other. The features of different scales are fused in a gradual mechanism, and the feature expression is aggregated through the spatial channel feature aggregation module (SCAM). A large number of experiments show that the MSNP-Net proposed in this paper has better completeness and lower false alarm rate for the detection and localization of spliced images. It can achieve more refined detection results, and has good stability and robustness. Significantly outperforms other state-of-the-art methods. Our source code is available at https://github.com/Swag-Jiang/MSNP-Net. • The multi-resolution feature extraction branch extracts local and global features. • The multi-scale noise-guided branch captures subtle tampering artifacts. • The dual-domain progressive fusion module enhances the information interaction. [ABSTRACT FROM AUTHOR]

LiDAR-only 3D detection methods struggle with the sparsity of point clouds. To overcome this issue, multi-modal methods have been proposed, but their fusion is a challenge due to the heterogeneous representation of images and point clouds. This paper proposes a novel multi-modal framework, Homogeneous Sparse Fusion (HS-Fusion), which generates pseudo point clouds from depth completion. The proposed framework introduces a 3D foreground-aware middle extractor that efficiently extracts high-responding foreground features from sparse point cloud data. This module can be integrated into existing sparse convolutional neural networks. Furthermore, the proposed homogeneous attentive fusion enables cross-modality consistency fusion. Finally, the proposed HS-Fusion can simultaneously combine 2D image features and 3D geometric features of pseudo point clouds using multi-representation feature extraction. The proposed network has been found to attain better performance on the 3D object detection benchmarks. In particular, the proposed model demonstrates a 4.02% improvement in accuracy compared to the pure model. Moreover, its inference speed surpasses that of other models, thus further validating the efficacy of HS-Fusion. • LiDAR-based 3D detection faces point cloud sparsity challenges. • A novel Homogeneous Sparse Fusion multi-modal approach is introduced. • Homogeneous Sparse Fusion adaptively extracts foreground features. • Cross-modality consistency fusion is achieved with multi-representation. • Homogeneous Sparse Fusion excels in accuracy and speed. [ABSTRACT FROM AUTHOR]

Several deep learning algorithms are currently focused on object detection in adverse weather scenarios for autonomous driving systems. However, these algorithms face challenges in real-time scenarios, leading to a reduction in detection accuracy. To tackle these issues, this paper introduces a lightweight object detection model named Self Driving Cars You Only Look Once Adverse Weather Detection Network (SD-YOLO-AWDNet), derived from enhancements to the YOLOv5 algorithm. The model incorporates four progressive improvement levels within the YOLOv5 framework. This includes integrating C3Ghost and GhostConv modules in the backbone to enhance detection speed by reducing computational overhead during feature extraction. To address potential accuracy issues arising from these modules, Depthwise-Separable Dilated Convolutions (DSDC) are introduced, striking a balance between accuracy and parameter reduction. The model further incorporates a Coordinate Attention (CA) module in the GhostBottleneck to enhance feature extraction and eliminate unnecessary features, improving precision in object detection. Additionally, a novel "Focal Distribution Loss" replaces CIoU Loss, accelerating bounding box regression and loss reduction. Test dataset experiments demonstrate that SD-YOLO-AWDNet outperforms YOLOv5 with a 54% decrease in FLOPs, a 52.53% decrease in model parameters, a 2.24% increase in mAP, and a threefold improvement in detection speed. [ABSTRACT FROM AUTHOR]

Pixel prediction is an important issue in the field of reversible data hiding. Neural networks are gradually used to improve the accuracy of pixel prediction owing to their excellent performance. However, current neural network-based pixel predictors are designed for natural images and do not consider the characteristics of medical images. Therefore, in this paper, we propose a dual-branch neural network-based reversible data hiding scheme for medical images. Detailedly, considering the characteristics of medical images, in which complex and smooth regions are more clearly distinguished, we present a clustering method to classify pixels into three classes according to their complexities, and generate masks to assist pixel prediction. Then, in the prediction stage, a dual-branch neural network-based pixel predictor is designed to extract unique and shared features, and a convolutional block attention module is used to optimize the extracted features. Finally, in the embedding stage, considering the characteristics of region of interest (ROI) and region of non-interest (NROI) in medical images, we design a class-based embedding algorithm, which can prioritize embedding data into NROI with low complexity and then sequentially into low texture complexity region and high texture complexity region of ROI. Experimental results show that our scheme can achieve better performance of pixel prediction and data embedding than existing state-of-the-art works. • A novel neural network-based PEE-RDH scheme for medical images is proposed. • We design an efficient neural network-based pixel predictor to obtain sharper PEH. • A class-based embedding strategy is exploited to reduce ROI distortion. • Unique features of different classes of pixels are considered in feature extraction. • Our scheme achieves better prediction accuracy and embedding performance. [ABSTRACT FROM AUTHOR]

Traffic congestion recognition is essential for road traffic condition monitoring and improving transportation operation efficiency. Recent works have proposed using computer vision to develop cost-effective traffic congestion detection systems. This paper proposed a custom framework for recognizing traffic congestion using a trained advanced CNNs model on highway surveillance images. The proposed CNNs model uses an attention mechanism that significantly helps the network performing better. Specifically, a spatial attention module is proposed to aggregate model features and increase representational power. Exploring the optimal model choice for edge device deployment: MobileNetV2 is considered the best model choice for this task due to its lowest FLOPs and params. Furthermore, the accuracy of MobileNetV2 with attention module on the test set exceeds that of several advanced CNNs models such as VGG16, GoogLeNet, ResNet50, Efficientnetb0, DenseNet121. For a selected dataset of real-world traffic congestion images, test accuracy of 98.58% strongly supports the model's efficiency in recognizing traffic congestion rendering image samples. In addition, proposed model also achieves 98.82% on the UCSD dataset, which as same as sort-of-the-art method. In summary, we have developed a congestion detection framework equipped with spatial attention module. Extensive experimental validation has demonstrated that using MobileNetV2 as the backbone network achieves state-of-the-art performance. The framework also uses the Grad-CAM approach for visual interpretability and localization of traffic congestion in surveillance images. Judging from the results of Grad-CAM, attention-integrated models can localize traffic congestion in surveillance images better. Our model is developed under the Pytorch framework using python3.8. • A deep learning framework for recognizing traffic congestion in real-world images. • A spatial attention module is proposed to extract features and improve accuracy. • Grad-CAM is used to highlight regions of the image. • High recall supports the reliability of model for congestion recognition task. [ABSTRACT FROM AUTHOR]

There is a close biological relationship between cats and humans. Emotion recognition of domestic cat meows plays an essential role in human–animal relationships, animal–environment relationships, and animal conservation. This study proposed a deep learning network named JL-TFMSFNet. It jointly learns the time–frequency domain information and multi-scale features to recognize the sound emotions of domestic cats. Firstly, the Mel feature of the domestic cat's sound was extracted based on the Mel filter, using the proposed Multi-scale Feature Extraction Module (MFEM) to extract weighting information to generate a deeper one. Secondly, the semantic information in the spectrogram was learned from the time–frequency domain using the Time–Frequency Attention Mechanism (TFAM) proposed in the backbone network. Then, the Diverse Branch Block (DBB) module with the different receptive fields was added to improve the recognition performance. Finally, validation experiments were conducted on the self-built dataset Cat_Emotion_Sound (CES) nd the public dataset Urbansound8K. The experiments demonstrated that JL-TFMSFNet outperformed the state-of-the-art sound classification models. The average accuracies of the proposed model in this paper on these two datasets are 94.43% and 96.14%, and the F1-score are 94.01% and 95.65%. • Constructed the dataset Cat_Emotion_Sound with five emotions of domestic cats. • Proposed a four-scale feature extraction module for depth feature extraction. • Combined time–frequency attention information and the multi-branch structure. • Introduced a model that joint learning. [ABSTRACT FROM AUTHOR]