Effective Feature Extraction Research Articles

Processing and Integration of Multimodal Image Data Supporting the Detection of Behaviors Related to Reduced Concentration Level of Motor Vehicle Users

This paper introduces a comprehensive framework for the detection of behaviors indicative of reduced concentration levels among motor vehicle operators, leveraging multimodal image data. By integrating dedicated deep learning models, our approach systematically analyzes RGB images, depth maps, and thermal imagery to identify driver drowsiness and distraction signs. Our novel contribution includes utilizing state-of-the-art convolutional neural networks (CNNs) and bidirectional long short-term memory (Bi-LSTM) networks for effective feature extraction and classification across diverse distraction scenarios. Additionally, we explore various data fusion techniques, demonstrating their impact on improving detection accuracy. The significance of this work lies in its potential to enhance road safety by providing more reliable and efficient tools for the real-time monitoring of driver attentiveness, thereby reducing the risk of accidents caused by distraction and fatigue. The proposed methods are thoroughly evaluated using a multimodal benchmark dataset, with results showing their substantial capabilities leading to the development of safety-enhancing technologies for vehicular environments. The primary challenge addressed in this study is the detection of driver states not relying on the lighting conditions. Our solution employs multimodal data integration, encompassing RGB, thermal, and depth images, to ensure robust and accurate monitoring regardless of external lighting variations

PGCNMDA: Learning node representations along paths with graph convolutional network for predicting miRNA-disease associations

Identifying miRNA-disease associations (MDAs) is crucial for improving the diagnosis and treatment of various diseases. However, biological experiments can be time-consuming and expensive. To overcome these challenges, computational approaches have been developed, with Graph Convolutional Network (GCN) showing promising results in MDA prediction. The success of GCN-based methods relies on learning a meaningful spatial operator to extract effective node feature representations. To enhance the inference of MDAs, we propose a novel method called PGCNMDA, which employs graph convolutional networks with a learning graph spatial operator from paths. This approach enables the generation of meaningful spatial convolutions from paths in GCN, leading to improved prediction performance. On HMDD v2.0, PGCNMDA obtains a mean AUC of 0.9229 and an AUPRC of 0.9206 under 5-fold cross-validation (5-CV), and a mean AUC of 0.9235 and an AUPRC of 0.9212 under 10-fold cross-validation (10-CV), respectively. Additionally, the AUC of PGCNMDA also reaches 0.9238 under global leave-one-out cross-validation (GLOOCV). On HMDD v3.2, PGCNMDA obtains a mean AUC of 0.9413 and an AUPRC of 0.9417 under 5-CV, and a mean AUC of 0.9419 and an AUPRC of 0.9425 under 10-CV, respectively. Furthermore, the AUC of PGCNMDA also reaches 0.9415 under GLOOCV. The results show that PGCNMDA is superior to other compared methods. In addition, the case studies on pancreatic neoplasms, thyroid neoplasms and leukemia show that 50, 50 and 48 of the top 50 predicted miRNAs linked to these diseases are confirmed, respectively. It further validates the effectiveness and feasibility of PGCNMDA in practical applications.

GAN-BodyPose: Real-time 3D human body pose data key point detection and quality assessment assisted by generative adversarial network

With the rapid advancement of deep learning and computer vision, these technologies are becoming increasingly vital in areas like virtual reality, medical diagnosis, and sports training. Existing methods for real-time 3D human body pose keypoint detection and quality assessment face significant challenges such as insufficient detection accuracy, low computational efficiency, and high data quality requirements. To address these challenges, we propose an innovative solution, GAN-BodyPose. This approach integrates 3D convolutional neural networks, self-attention mechanisms, and generative adversarial networks to deliver efficient and accurate detection and assessment in real time. The GAN-BodyPose framework combines 3D-CNN and self-attention for effective feature extraction and keypoint detection, enhanced further by generative adversarial networks for superior data quality and accuracy. Our extensive evaluations using a large-scale 3D human body pose dataset demonstrated that GAN-BodyPose outperforms traditional methods, showing improvements in processing speed (15% faster), accuracy in terms of Mean Per Joint Position Error (reduced by approximately 2.2%), and an Area Under the Curve (AUC) score increased by approximately 9.5% compared to HR-Net and other datasets. Additionally, it achieves lower Floating-Point Operations (FLOPs) by about 9.3%, indicating more efficient computational performance. These advancements underline the potential of our approach to significantly enhance user experiences in virtual reality, motion capture, and other real-time applications. The successful application of GAN-BodyPose promises greater efficiency and precision in fields ranging from game development to medical diagnostics, and robust support for human-computer interaction and gesture recognition. This research represents a substantial contribution to deep learning applications in robot control, decision-making, and broadens the research foundation in these domains.

Ethereum Phishing Scam Detection Based on Data Augmentation Method and Hybrid Graph Neural Network Model.

The rapid advancement of blockchain technology has fueled the prosperity of the cryptocurrency market. Unfortunately, it has also facilitated certain criminal activities, particularly the increasing issue of phishing scams on blockchain platforms such as Ethereum. Consequently, developing an efficient phishing detection system is critical for ensuring the security and reliability of cryptocurrency transactions. However, existing methods have shortcomings in dealing with sample imbalance and effective feature extraction. To address these issues, this study proposes an Ethereum phishing scam detection method based on DA-HGNN (Data Augmentation Method and Hybrid Graph Neural Network Model), validated by real Ethereum datasets to prove its effectiveness. Initially, basic node features consisting of 11 attributes were designed. This study applied a sliding window sampling method based on node transactions for data augmentation. Since phishing nodes often initiate numerous transactions, the augmented samples tended to balance. Subsequently, the Temporal Features Extraction Module employed Conv1D (One-Dimensional Convolutional neural network) and GRU-MHA (GRU-Multi-Head Attention) models to uncover intrinsic relationships between features from the time sequences and to mine adequate local features, culminating in the extraction of temporal features. The GAE (Graph Autoencoder) concept was then leveraged, with SAGEConv (Graph SAGE Convolution) as the encoder. In the SAGEConv reconstruction module, by reconstructing the relationships between transaction graph nodes, the structural features of the nodes were learned, obtaining reconstructed node embedding representations. Ultimately, phishing fraud nodes were further identified by integrating temporal features, basic features, and embedding representations. A real Ethereum dataset was collected for evaluation, and the DA-HGNN model achieved an AUC-ROC (Area Under the Receiver Operating Characteristic Curve) of 0.994, a Recall of 0.995, and an F1-score of 0.994, outperforming existing methods and baseline models.

A Spatio-temporal Graph Transformer driven model for recognizing fine-grained data human activity

Human activity recognition(HAR) is an important research focus in ubiquitous computing. It has been widely applied in various domains, such as smart homes, healthcare assistance, and sports training. Accurate human activity recognition directly impacts the performance of downstream tasks. Existing methods for human activity recognition primarily rely on extracting temporal or spatial features from the data. These features are used for the task of human activity recognition. The existing methods mainly use CNN or RNN models to model the Euclidean correlations among spatially adjacent sensors or channels. However, non-Euclidean pairwise correlations among all sensors or channels are even critical for accurate classification, which has been ignored by the existing methods. In this paper, we incorporate fine-grained spatial structural data into the model to overcome these limitations. A novel deep learning model for human activity recognition is proposed, which is called the fine-grained data-oriented Spatial-Temporal Graph Transformer network (STGT). The introduction of the STGT model can eliminate the limitations of existing spatial feature extraction methods by leveraging a novel data organization approach proposed in this study. This model enhances the spatial features within the time feature extraction module for effective spatio-temporal feature extraction. We conducted experiments on four large-scale real-world HAR datasets to evaluate its performance. The experimental results demonstrate the superiority of our method over state-of-the-art approaches.

Adaptive feature extraction of underwater target signal based on mathematical morphology feature enhancement

This paper proposes an adaptive feature extraction method based on mathematical morphology enhancement to extract effective and stable features under strong ocean noise. Firstly, the traditional mathematical morphology method is improved and a new mathematical morphology filtering method is proposed for feature enhancement of underwater target signals. Secondly, the artificial fish swarm algorithm (AFSA) is improved using weighted power spectral kurtosis (WPSK) to achieve parameter adaptivity. Five measured underwater target signals are used to validate the extracted method, and cosine similarity (CS), signal to noise ratio (SNR), and refined composite multiscale fluctuation dispersion entropy (RCMFDE) are used as evaluation metrics to assess the results and verify the effectiveness of the proposed method. Compared with average filtering (AVGF) and closing opening filtering (COF), the proposed method shows better capability in adaptive feature extraction. Therefore, the proposed mathematical morphological filtering can enhance underwater target features and is very important for adaptive feature extraction of underwater targets.

Representation Learning Based on Vision Transformer

In recent years, with the rapid development of information technology, the volume of image data has grown exponentially. However, these datasets typically contain a large amount of redundant information. To extract effective features and reduce redundancy from images, a representation learning method based on the Vision Transformer (ViT) has been proposed, and to our best knowledge, Transformer was first applied to zero-shot learning (ZSL). The method adopts a symmetric encoder–decoder structure, where the encoder incorporates Multi-Head Self-Attention (MSA) mechanism of ViT to reduce the dimensionality of image features, eliminate redundant information, and decrease computational burden. Consequently, it effectively extracts features, and the decoder is utilized for reconstructing image data. We evaluated the representation learning capability of the proposed method in various tasks, including data visualization, image reconstruction, face recognition, and ZSL. By comparing with state-of-the-art representation learning methods, the outstanding results obtained validate the effectiveness of this method in the field of representation learning.

Feature extraction method for ultrasonic pipeline defects based on fractional-order VMD

ABSTRACT Ensuring the safety of pipeline transportation is vital for societal well-being. Traditional methods for identifying defects in steel pipelines through ultrasonic echo signal pattern recognition often fail to extract comprehensive and effective features crucial for accurate defect detection. This study introduces an innovative feature extraction method employing a fractional Fourier transform variational modal decomposition (FRFT-VMD), hereafter referred to as the fractional-order VMD algorithm. This method utilises the fourth-order central moment and envelope entropy to optimise several key parameters: the fractional order in the Fourier transform, the number of decomposition layers, and the penalty factor in variational modal decomposition. To evaluate the effectiveness of the proposed method, ultrasonic echo signals from both finite element simulations and experimental platforms were analysed using the FRFT-VMD technique. The features extracted were then classified using a Least Squares Support Vector Machine (LSSVM) to determine defect depths. The results show a recognition accuracy of 95.2% in simulated signals and 89.1% in experimentally measured signals across various defect depths, indicating a significant improvement over existing feature extraction methodologies. The fractional-order VMD algorithm proves to be superior in extracting features that enhance the accuracy of defect identification in steel pipelines.

Hybrid model based on similar power extraction and improved temporal convolutional network for probabilistic wind power forecasting

Accurate wind power generation forecasting is of great significance to improve the operation of power system. Probabilistic forecasting has a higher application value in power grid because it can provide more abundant forecasting information than deterministic forecasting. In addition, multi-step forecasting can provide forecasting results in a longer time range, so that decision makers can make longer-term planning and strategic arrangements. In this paper, we propose a novel multi-step improved temporal convolutional network based on quadratic spline quantile function (MITCN-QSQF) for probabilistic wind power forecasting. First, we combine maximum information coefficient, Gaussian similarity and adaptive resample to propose an effective similar power generation feature extraction method (MGR) for power generation. Then the temporal convolutional network is improved to construct the multi-step time series forecasting model MITCN. By combining the proposed model and the powerful probabilistic forecasting method quadratic spline quantile function (QSQF), high-quality probabilistic forecasting of wind power is achieved. Through comprehensive simulations on an open-source dataset, the superiority and efficiency of the proposed method are verified. Compared with some advanced benchmarks, the proposed model can obtain more accurate deterministic and probabilistic forecasting results.

Exploration on multimodal data recognition method for Internet of Things based on deep learning

With the rapid development of Internet of Things (IoT) technology, a large amount of sensor data, images, voice, and other data are being widely used, bringing new opportunities for intelligent and cross-domain information fusion. Effective feature extraction and accurate recognition remain urgent issues to be addressed. This article explores the application of deep learning (DL) in multimodal data recognition methods of the IoT and proposes path optimization for multimodal data recognition methods of the IoT under DL. This article also provides in-depth analysis and discussion on the optimization of multimodal data recognition models based on DL, as well as specific measures for optimizing the path of multimodal data recognition based on DL. In this paper, the long short-term memory (LSTM) technology is introduced, and the LSTM technology is used to optimize the multi-modal data recognition method. It can be seen from the comparison that the processing efficiency of data analysis, information fusion, speech recognition, and emotion analysis of the multimodal data recognition method optimized by LSTM technology is 0.29, 0.35, 0.31, and 0.24 higher, respectively, than that of data analysis, information fusion, speech recognition, and emotion analysis before optimization. Introducing DL methods in multimodal data recognition of the IoT can effectively improve the effectiveness of data recognition and fusion and achieve higher levels of recognition for speech recognition and sentiment analysis.

Detecting tumors in medical images using segmentation and feature extraction techniques

Tumors rank among the most prevalent and lethal diseases worldwide, emphasizing the paramount importance of early detection in mitigating mortality rates. There has been a tremendous growth in medical image segmentation and classification for diagnosing cancer. This led to the various developments in medical research that help in diagnosing types of cancers and other human medical problems. This work focuses on performing effective segmentation procedures and feature extraction by a proposed hybrid intelligent technique for detecting tumors in the medical images. Additionally, the research addresses the design aspects of feature extraction, image classification, and texture features. These features are reduced by employing feature selection. Finally, experiments were conducted to compare tumor detection in medical images using the proposed algorithm with other existing algorithms. The results demonstrated promising achievements in terms of their evaluation metrics.

Comparative Study of sEMG Feature Evaluation Methods Based on the Hand Gesture Classification Performance.

Effective feature extraction and selection are crucial for the accurate classification and prediction of hand gestures based on electromyographic signals. In this paper, we systematically compare six filter and wrapper feature evaluation methods and investigate their respective impacts on the accuracy of gesture recognition. The investigation is based on several benchmark datasets and one real hand gesture dataset, including 15 hand force exercises collected from 14 healthy subjects using eight commercial sEMG sensors. A total of 37 time- and frequency-domain features were extracted from each sEMG channel. The benchmark dataset revealed that the minimum Redundancy Maximum Relevance (mRMR) feature evaluation method had the poorest performance, resulting in a decrease in classification accuracy. However, the RFE method demonstrated the potential to enhance classification accuracy across most of the datasets. It selected a feature subset comprising 65 features, which led to an accuracy of 97.14%. The Mutual Information (MI) method selected 200 features to reach an accuracy of 97.38%. The Feature Importance (FI) method reached a higher accuracy of 97.62% but selected 140 features. Further investigations have shown that selecting 65 and 75 features with the RFE methods led to an identical accuracy of 97.14%. A thorough examination of the selected features revealed the potential for three additional features from three specific sensors to enhance the classification accuracy to 97.38%. These results highlight the significance of employing an appropriate feature selection method to significantly reduce the number of necessary features while maintaining classification accuracy. They also underscore the necessity for further analysis and refinement to achieve optimal solutions.

Effective features extraction and selection for hand gesture recognition using sEMG signal

Effective features extraction and selection for hand gesture recognition using sEMG signal

Prediction of oxygen content in flue gas of power plant with stacked target-enhanced autoencoder and attention-based LSTM

Accurate and robust measurement of the oxygen content in the flue gas of power plant is significant to optimal the coal combustion process. Conventional hard measurement algorithms have low accuracy, short lifetime and high cost. Therefore, it is urgent to establish a precise and reliable soft measurement model. However, how to extract effective feature representations from complex process data is still the difficult and hot spot in the soft measurement field. The paper develops a novel soft measurement model, which integrates stacked target-enhanced autoencoder (STAE) with attention mechanism-based long short-term memory (ALSTM) in view of the above issues. The data cleaning strategy is firstly designed to filter the redundant data from the original data. Then STAE is developed to extract the useful features, which introduces target output into each autoencoder to ensure the feature representation is output-related and stacks several autoencoder units to learn high-level information. The extracted hierarchical features can not only represent the input data, but also have high correlation with the target output. A three-layer-based attention network is designed to allocate different weights on the different levels’ features. Finally, due to the dynamic characteristic of the problem, a LSTM-based prediction network is dexterously designed to learn the potential relation from extracted features to target output, i.e., oxygen content. Numerous real-word data is collected from a 300 MW power plant and extensive experiments are conducted. The maximum prediction errors of A and B are 0.42 and 0.41. The 95.1 % prediction errors of A side are within [−0.2, 0.2], and the 96.2 % prediction errors of B side are within [−0.2, 0.2].

Dimensionality Reduction in Structural Health Monitoring: A Case Study on Damaged Wind Turbine Blades

AbstractIn recent years, advances in structural health monitoring (SHM) of composite materials have been observed, driven by the adoption of intelligent diagnostic techniques. Vibration signals, which embed valuable structural health information, have been increasingly utilized in SHM. However, the high dimensionality of this data has necessitated significant computational resources and has made the differentiation between health states more complex. Highlighting the need for effective feature extraction and dimensionality reduction (DR), a study is conducted. In this research, advanced DR techniques, i.e., principal component analysis (PCA), locally linear embedding (LLE), t‐distributed stochastic neighbors embedding (t‐SNE), and uniform manifold approximation and projection (UMAP), are applied to an experimental dataset from a wind turbine blade under various health and environmental conditions (controlled‐environment vibration signals). The extracted features are then processed for the classification phase, and it is found that UMAP provides the best performance, albeit with a slightly increased computational demand. The findings from this research offer invaluable insights for researchers and engineers, assisting in the selection of the most appropriate DR method for SHM, and weighing up accuracy against computational time.

No reference image quality assessment using gabor convolutional neural networks

With the emergence of image capturing devices and increase in usage of internet, massive volume of network data was occupied with digital images. For efficient data transmission, the image may undergo several processing units in from the point it captures to the display/storage device. It may result in loss of the perceptual image quality. Therefore, it is necessary to estimate the image quality to measure the quality of experience. It was found that the convolutional neural networks serve as potential tool for effective feature extraction in many image processing applications. Particularly, with the first layer as Gabor filters, the robustness of the network can be reinforced with learnable Gabor parameters. This paper proposes Gabor Convolutional Neural Network method for No-Reference Image Quality Assessment. Their well-defined spatial structured filters are promising in extracting quality features from the local patches and maps them to perceptual quality scores. Our proposed architecture was tested over synthetic and authentic databases such as LIVE, TID2013, CSIQ, LIVE-MD, MDID2016, LIVE Wild and KoNiQ-10k. The proposed approach was also tested on the Waterloo 3D phase-II database, which contains high-resolution images of both the eyes individually with their respective DMOS scores. The proposed approach out performs over LIVE-MD and LIVE Wild and competes with existing algorithm over other databases.

Enhancing Stock Trading Strategies: Integrating Discrete Wavelet Transformation with Deep Q-Network

The prediction of price trends in the stock market has always been a hot research topic in the financial field. However, due to the high instability and volatility of stock prices, it is very difficult to accurately predict stock trends. How to remove the noise of stock data, extract effective features, and pursue maximum value returns has always been a challenge. This paper proposes a hybrid model (DWT-DQN) that combines discrete wavelet transform with deep reinforcement learning to improve the accuracy and return rate of stock predictions. First, the model captures price fluctuation information on different scales by performing discrete wavelet transformation on the difference between long- and short-term moving averages of stock prices, and well extracts the changing characteristics of stock price data in the time domain and frequency domain. Then the feature data are input into the built DQN network for model training. The network can select the optimal trading action based on market status and historical experience and returns. At the same time, during the data sampling process, an attention mechanism is introduced to allow the model to further learn in the direction of maximizing returns. Through testing and verification on SSEC, HSI, NDX and SPX data sets, experiments show that the hybrid model proposed in this paper has excellent performance in terms of accuracy and return rate.

Hyperspectral Image Classification Based on Two-Branch Multiscale Spatial Spectral Feature Fusion with Self-Attention Mechanisms

In recent years, the use of deep neural network in effective network feature extraction and the design of efficient and high-precision hyperspectral image classification algorithms has gradually become a research hotspot for scholars. However, due to the difficulty of obtaining hyperspectral images and the high cost of annotation, the training samples are very limited. In order to cope with the small sample problem, researchers often deepen the network model and use the attention mechanism to extract features; however, as the network model continues to deepen, the gradient disappears, the feature extraction ability is insufficient, and the computational cost is high. Therefore, how to make full use of the spectral and spatial information in limited samples has gradually become a difficult problem. In order to cope with such problems, this paper proposes two-branch multiscale spatial–spectral feature aggregation with a self-attention mechanism for a hyperspectral image classification model (FHDANet); the model constructs a dense two-branch pyramid structure, which can achieve the high efficiency extraction of joint spatial–spectral feature information and spectral feature information, reduce feature loss to a large extent, and strengthen the model’s ability to extract contextual information. A channel–space attention module, ECBAM, is proposed, which greatly improves the extraction ability of the model for salient features, and a spatial information extraction module based on the deep feature fusion strategy HLDFF is proposed, which fully strengthens feature reusability and mitigates the feature loss problem brought about by the deepening of the model. Compared with five hyperspectral image classification algorithms, SVM, SSRN, A2S2K-ResNet, HyBridSN, SSDGL, RSSGL and LANet, this method significantly improves the classification performance on four representative datasets. Experiments have demonstrated that FHDANet can better extract and utilise the spatial and spectral information in hyperspectral images with excellent classification performance under small sample conditions.

Fine grained food image recognition based on swin transformer

Fine-grained food image recognition is an important research direction in the field of computer vision and machine learning. However, fine-grained food image recognition faces huge challenges when dealing with foods that vary greatly in shape but belong to the same category or subcategories of that food. To improve this problem, this paper proposes a deep convolution module for obtaining local enhanced feature representation and combines it with the global feature representation obtained from Swin Transformer for deep residual, to obtain a deeper enhanced feature representation. An end-to-end fine-grained food universal classifier was also proposed, which can more accurately extract effective feature information from enhanced feature representations and achieve accurate recognition. Our approach can accurately handle foods with widely different shapes but belonging to the same category and is expected to help people better manage their diet and improve their health. Our models were trained and verified on the public fine-grained food datasets Foodx-251 and UEC Food-256 respectively, where the accuracy of the method on the validation set is 81.07% and 82.77% respectively. Compared with other state-of-the-art self-supervised methods, the method proposed in this paper exhibits higher accuracy in fine-grained food image recognition tasks.

Influence of VR-Assisted College Dance on College Students' Physical and Mental Health and Comprehensive Quality

With the development and the popularization of sports dance, sports dance teaching has become a required elective course in universities. Sports dance can not only improve students' comprehensive quality, but also affect college students' healthy psychology. The use of VR (Virtual Reality) technology in dance education will definitely develop and promote dance education. This paper studies an effective feature extraction method for the characteristics of dance movements based on VR. The edge features of all video images in each segment are accumulated into one image, and the directional gradient histogram features are extracted from it. The results show that compared with the current robust regression method and cascade regression method, our method has higher positioning accuracy on the pollution test set, and more than 75% of the sample errors in this method are within 0.1. This also verifies the effectiveness of this motion recognition algorithm for dance motion recognition. Dance can effectively resist the psychological barriers of college students and improve their comprehensive quality.