Most Download

Published in last 1 year| In last 2 years| In last 3 years| All| Most Downloaded in Recent Month | Most Downloaded in Recent Year|

Published in last 1 year

Please wait a minute...


For Selected: Download Citations EndNote Ris BibTeX Toggle Thumbnails

Select

Review of Steganalysis for Digital Images WANG Zichi, LI Bin, FENG Guorui, ZHANG Xinpeng Journal of Applied Sciences    2024, 42 (5): 723-732.   DOI: 10.3969/j.issn.0255-8297.2024.05.001 Abstract （539）      PDF（pc） （2695KB）（503）       Knowledge map    Save Digital steganography plays a crucial role in securely transmitting confidential data by concealing it within common multimedia, such as images, videos, and audio, to facilitate covert communication. To discover the covert communication of steganography, the technique of steganalysis can be employed. Steganalysis determines whether a given multimedia object contains secret data according to the statistical anomaly of stego data caused by steganography. In recent years, both steganography and steganalysis have made significant progress and development in their mutual confrontation, particularly in the context of the growing prevalence of digital images on social networks. Focusing on digital images, this paper sorts out the development of steganalysis in the past decade, and reviews the traditional steganalysis and deep learning based-steganalysis. Then, the limitations of each approach are discussed. Finally, the study offers insights into the prospective development trends in steganalysis. Reference | Related Articles | Metrics | Comments（0）

Select

Cross-Modal Person Re-identification Driven by Cross-Channel Interactive Attention Mechanism in Dual-Stream Networks HE Lei, LI Fengyong, QIN Chuan Journal of Applied Sciences    2024, 42 (5): 884-892.   DOI: 10.3969/j.issn.0255-8297.2024.05.014 Abstract （250）      PDF（pc） （4390KB）（485）       Knowledge map    Save Existing cross-modal person re-identification methods often fail to take into account the difference of target person between modes and within modes, making it difficult to further improve the retrieval accuracy. To solve this problem, this paper introduces the cross-channel interaction attention mechanism to enhance the robust extraction of person features, effectively suppresses the extraction of irrelevant features and achieves more discriminative feature expression. Furthermore, hetero-center triplet loss, triplet loss and identity loss are combined for supervised learning, effectively integrating the intermodal and intra-class differences in person features. Experimental results demonstrate the effectiveness of the proposed method, which outperforms seven existing methods on two standard datasets, RegDB and SYSU-MM01. Reference | Related Articles | Metrics | Comments（0）

Select

News Recommendation Algorithm Incorporating Headline Sentiment and Topic Characteristics AI Jun, HONG Xingqi Journal of Applied Sciences    2024, 42 (5): 810-822.   DOI: 10.3969/j.issn.0255-8297.2024.05.008 Abstract （251）      PDF（pc） （1748KB）（379）       Knowledge map    Save Traditional lexicon-based news recommendation algorithms often ignore the emotional nuances present in words beyond the confines of the dictionary. This oversight can lead to issues such as diminished prediction accuracy and subpar sorting performance. To address these challenges, this paper introduces a heuristic approach to deduce the sentiment of unfamiliar words and devises a news recommendation algorithm to verify its feasibility. A tripartite graph model is constructed to propagate sentiment from a sentiment dictionary to individual words and obtain the headline sentiment. In addition, the bag-of-words model is used to extract topic features from the headlines. The sentiment similarity and topic similarity between headlines are calculated, consolidating these into a comprehensive similarity evaluation index. The news with higher similarity to the target news is then selected as the neighbor. The algorithm predicts the hourly average click volume of the target news by considering the hourly average click volume of neighbors, treating this as the predicted score for the target news. Finally, users receive a selection of high-scoring news articles. Validation using real data from NetEase News confirms the feasibility and effectiveness of our algorithm. Compared with other algorithms, our algorithm has shown improvements in the optimal accuracy of mean absolute error (MAE) by 2.2% to 3.4%, root mean square error (RMSE) by 2.3% to 2.9%, and the mean score of normalized discounted cumulative gain (NDCG) by 0.7% to 1.8%, respectively. Reference | Related Articles | Metrics | Comments（0）

Select

Auditable and Traceable Blockchain Privacy Protection Model under Zero-Knowledge Proof WU Meng, QI Yong Journal of Applied Sciences    2024, 42 (4): 598-612.   DOI: 10.3969/j.issn.0255-8297.2024.04.004 Abstract （262）      PDF（pc） （558KB）（373）       Knowledge map    Save In order to address the issues of sensitive data exposure due to shared ledgers among nodes in a blockchain network, alongside the inability to audit and trace encrypted privacy data, a blockchain privacy protection model based on directed graphs and zeroknowledge proofs has been proposed. This model extends the open-source Hyperledger Fabric framework and effectively inherits the features of Fabric. By encrypting on-chain transaction information and utilizing Pedersen commitments and Schnorr-type zero-knowledge proofs, it generates proofs of balance, traceability, asset ownership, and consistency to provide fast and verifiable privacy data audits. The model utilizes a directed graph structure to construct a transaction graph, thus achieving traceability of transaction information on the blockchain. Moreover, it generates proofs to validate the correctness of forward tracing transactions. Experimental results demonstrate that the proposed model achieves complete audit and traceability on Fabric at a cost of less than 10% throughput, outperforming existing related models. Reference | Related Articles | Metrics | Comments（0）

Select

Entity Relationship Extraction Framework Based on Pre-trained Large Language Model and Its Application WEI Wei, JIN Chenggong, YANG Long, ZHOU Mo, MENG Xiangzhu, FENG Hui Journal of Applied Sciences    2025, 43 (1): 20-34.   DOI: 10.3969/j.issn.0255-8297.2025.01.002 Abstract （362）      PDF（pc） （1446KB）（354）       Knowledge map    Save Entity relationship extraction is a crucial foundation for building large-scale knowledge graphs and domain-specific datasets. This paper proposes an entity relationship extraction framework based on pre-trained large language models (PLLM-RE) for relation extraction in circular economy policies. Within this framework, entity recognition of circular economy policy texts is performed based on the model RoBERTa. Subsequently, the bidirectional encoder representation from Transformers (BERT) is employed for entity relation extraction, facilitating the construction of a knowledge graph in the field of circular economic policies. Experimental results demonstrate the framework outperforms the baseline models including BiLSTM-ATT, PCNN, BERT and ALBERT in task of entity relationship extraction for circular economy policies. These findings validate the adaptability and superiority of the proposed framework, providing new ideas for information mining and policy analysis in the field of circular economy resources in the future. Reference | Related Articles | Metrics | Comments（0）

Select

A Low Light Image Enhancement Method Based on CRTNet JIANG Zetao, HUANG Jingfan, ZHU Wencai, HUANG Qinyang, JIN Xin Journal of Applied Sciences    2024, 42 (6): 934-946.   DOI: 10.3969/j.issn.0255-8297.2024.06.004 Abstract （526）      PDF（pc） （4085KB）（253）       Knowledge map    Save A low light image enhancement method based on color restoration transformer networks (CRTNet) is proposed to address the issue of image and color distortion in low light environments. This method combines channel attention and spatial attention mechanisms. CRTNet consists of a color attention module (CAM), a color map module (CMM), and a sequential enhancement structure. Firstly, CAM is divided into two parts: color channel attention module and color space attention module. Utilizing the global information capture capability of Transformer, the color channel attention module emphasizes meaningful color channels by assigning higher weights to generate channel attention vectors. The color space attention module uses a three-layer convolution structure, focuses on spatial details in high-dimensional space and generates spatial attention weight map. Secondly, CMM extracts high-dimensional image features through a linear fitting process, scaling and shifting these features in the 64D space across both channel and spatial dimensions to obtain global and detail image information. By combining with the original image features, it supplements the color, brightness, contrast, and detail information in the original image features to achieve color enhancement. Finally, a sequential enhancement structure is adopted to repeat CAM and CMM operations three times with the output of CMM serving as input, in order to fit higher-order function mappings and effectively enhance low light images. Experiments results and user studies on public datasets demonstrate that the proposed method outperforms existing approaches in quantitative measurement, detail and color restoration. Reference | Related Articles | Metrics | Comments（0）

Select

Benchmarking of Spiking Neural Networks and Performance Evaluation of Neuromorphic Training Frameworks HU Wangxin, CHENG Yingchao, HE Yulin, HUANG Zhexue, CAI Zhanchuan Journal of Applied Sciences    2025, 43 (1): 169-182.   DOI: 10.3969/j.issn.0255-8297.2025.01.012 Abstract （310）      PDF（pc） （805KB）（249）       Knowledge map    Save With the growing interest in spiking neural networks (SNNs), the development of open-source neuromorphic training frameworks has also accelerated. However, there is currently a lack of systematic guidelines for selecting these frameworks. To address this issue, this paper proposes a benchmarking method for SNNs based on image classification tasks. This method designs a convolutional neural network and a fully connected deep neural networks to evaluate two SNN training approaches: direct training with surrogate gradient backpropagation and conversion from artificial neural networks (ANNs) to SNNs. Based on the MNIST, Fashion-MNIST, and CIFAR-10 benchmark image datasets, the performance comparisons of various neuromorphic training frameworks are conducted using evaluation metrics such as training time and classification accuracy. Experimental results indicate that the neuromorphic training framework SpikingJelly outperforms others in terms of both training time and classification accuracy in direct SNN training, while the Lava framework achieves the highest classification accuracy in ANN-to-SNN conversion training. Reference | Related Articles | Metrics | Comments（0）

Select

A Multimodal Knowledge Graph Entity Alignment Method LIU Wei, XU Hui, LI Weimin Journal of Applied Sciences    2024, 42 (6): 1040-1051.   DOI: 10.3969/j.issn.0255-8297.2024.06.012 Abstract （785）      PDF（pc） （1304KB）（248）       Knowledge map    Save The fusion of multimodal knowledge graph requires addressing the entity alignment problem in knowledge fusion. In multimodal knowledge graph, multimodal attributes can provide key alignment information to improve entity alignment effectiveness. This paper proposes a method for entity alignment in multimodal knowledge graphs based on multimodal attribute embedding and graph attention network. First, the multimodal knowledge graph is divided into subgraphs according to image, text and graph structure information. Text and graph structure information are then extracted by graph attention network, while image information is extracted by visual geometry group (VGG) network. These multimodal attributes are embedded into vector space. Finally, the proposed method integrates the multimodal attributes and the graph structure of the subgraphs for alignment. Experimental results shows that the proposed model significantly improves performance, achieving increases of 10.64% on Hits@1, 5.60% on Hits@10, and 0.226 on MRR compared to four baseline models for entity alignment. Reference | Related Articles | Metrics | Comments（0）

Select

Video-Based Facial Feature Computation Methods WANG Yingxiao, YANG Yanhong, TAN Yunfeng Journal of Applied Sciences    2025, 43 (1): 137-153.   DOI: 10.3969/j.issn.0255-8297.2025.01.010 Abstract （202）      PDF（pc） （730KB）（239）       Knowledge map    Save This paper presents a review of research in video face recognition conducted over the past five years. It provides a comparative analysis of the facial feature computation methods, categorizing them into traditional approaches, deep learning techniques, and feature aggregation/fusion methods. Traditional feature extraction methods include linear and nonlinear approaches, while deep learning methods include spatial and temporal feature extraction techniques. Feature aggregation and fusion methods integrate multiple feature sources and fuse features from different time periods to improve recognition performance. At the end of each subsection, this paper also provides a unified analysis of the algorithms used in the literature, highlighting their advantages, evaluation metrics, and applications. Through this research, we aim to provide more reliable and efficient solutions for practical applications of video face recognition systems and promote further advancements in this field. Reference | Related Articles | Metrics | Comments（0）

Select

High Speed Demodulation System for FBG Current Sensor Based on Mode-Locked Laser WANG Hua, HE Qun, TAN Ruchao, WU Dong, FANG Yinuo, MA Yuehui, YAN Kaiquan, MOU Chengbo Journal of Applied Sciences    2024, 42 (6): 903-911.   DOI: 10.3969/j.issn.0255-8297.2024.06.001 Abstract （571）      PDF（pc） （2185KB）（225）       Knowledge map    Save The development goals of reliable, safe, economical, and efficient smart grid place higher demands on the detection rate of current parameters. This paper presents an experimental investigation of high-speed demodulation of fiber Bragg grating (FBG) current sensor, based on magnetostrictive effect and mode-locked laser multiplexing. For the first time, time-stretch dispersive Fourier transformation (TS-DFT) is combined with fiber current sensing techniques. FBG, fixed on the magnetostrictive material, detects the material strain caused by the magnetic field generated by the energized solenoid, enabling current sensing. TS-DFT maps the wavelength shift of FBG caused by stress to the time-domain delay shift in the reflected pulse, facilitating high-speed demodulation. The wavelength multiplexing of the two sensing FBGs is monitored in the current range of 0 to 4.5 A, achieving a demodulation rate of up to 69.6 MHz. This method has broad application prospects in the field of current or magnetic field sensing. Reference | Related Articles | Metrics | Comments（0）

Select

Emotion Recognition of EEG Using Subdomain Adaptation and Spatial-Temporal Learning TANG Yiheng, WANG Yongxiong, WANG Zhe, ZHANG Xiaoli Journal of Applied Sciences    2024, 42 (6): 1016-1026.   DOI: 10.3969/j.issn.0255-8297.2024.06.010 Abstract （675）      PDF（pc） （1384KB）（212）       Knowledge map    Save In cross-subject emotion recognition, there are significant differences in the distribution of electroencephalogram (EEG) samples among different subjects, and domain adaptation is commonly used to alleviate the differences. However, the differences in the EEG distribution across affective subdomains are ignored by global adaptation, which reduces the distinguishability of emotional features. Besides, EEG contains a number of electrodes, and subjects only reach the prospective emotion during part of stimuli. Learning the complex spatial information between channels and emphasizing critical EEG frames is essential. Hence, we propose a subdomain adaptation and spatial-temporal learning network for EEG-based emotion recognition. In the subdomain adaptation module, the difference loss in subdomains is reduced by minimizing intra-class differences and maximizing inter-class differences. A spatial-temporal feature extractor captures spatial correlations and temporal contexts, extracting discriminative emotional features. Subject-independent experiments conducted on the public DEAP dataset demonstrate the superior performance of the proposed method, achieving classification accuracies of 0.688 0 for arousal and 0.696 8 for valence, respectively. Reference | Related Articles | Metrics | Comments（0）

Select

Design of Planar Scanning System for Electromagnetic Near-Field Testing JIA Hongchuan, CHENG Xin, WAN Fayu, RAVELO Blaise Journal of Applied Sciences    2024, 42 (5): 893-902.   DOI: 10.3969/j.issn.0255-8297.2024.05.015 Abstract （240）      PDF（pc） （6118KB）（200）       Knowledge map    Save This paper develops a design of near-field scanning system for measuring the electromagnetic (EM) field radiated by electronic devices for EM compatibility. Firstly, a magnetic field probe implemented on four-layer printed circuit board (PCB) structure is designed, with a working frequency of up to 12 GHz and a spatial resolution of 2 mm. The simulation results match well with the experimental measurements, and the probe is calibrated accordingly. Secondly, automation of the near-field scanning is achieved by designing a position machine using LabVIEW, with the STM32 serving as the motion control core. The STM32 receives the positioning data through the serial port, and controls the stepper motor to drive the probe for fixed-point scanning. The position machine communicates with the vector network analyzer through the local area network to read and save data. Finally, the data is visualized by the position machine. The results are calibrated upon completion of the scanning process, producing a real-time visualization of the field distribution map of the tested object. The measured field strength results show good agreement with electromagnetic simulation results, demonstrating the system’s suitability for analyzing electromagnetic coupling paths and transitioning between near-field and farfield regions. Reference | Related Articles | Metrics | Comments（0）

Select

Autonomous Driving Algorithm Based on Meta-Learning and Reinforcement Learning JIN Yanliang, FAN Baorong, GAO Yuan, WANG Xiaoyong, GU Chenjie Journal of Applied Sciences    2024, 42 (5): 795-809.   DOI: 10.3969/j.issn.0255-8297.2024.05.007 Abstract （277）      PDF（pc） （5419KB）（180）       Knowledge map    Save To address the problems of convergence difficulty, unsatisfactory training effect and poor generalization performance of autonomous driving algorithms based on reinforcement learning, an autonomous driving system based on meta-learning and reinforcement learning is proposed in this paper. The system first combines variational auto encoder (VAE) with Wasserstein generative adversarial network incorporating gradient penalty (WGAN-GP) to form the VWG (VAE-WGAN-GP) model, which improves the quality of extracted feature. Then, the meta learning algorithm Reptile is used to train the VWG feature extraction model, yielding the MVWG (Meta-VWG) feature extraction model. This approach accelerates the training speed. Finally, the feature extraction model is combined with the proximal policy optimization (PPO) decision algorithm, and the reward function in the PPO algorithm is refined to enhance the convergence speed of the decision model, resulting in the MVWG-PPO autonomous driving model. Experimental results show that compared with VAE, VW (VAE-WGAN) and VWG benchmark models, the MVWG feature extraction model proposed in this paper reduces reconstruction loss by 60.82%, 44.73%, and 29.09%, respectively. The convergence rate increases approximately fivefold, achieving clearer reconstructed images and superior performance in automatic driving tasks. It can provide higher-quality feature information for autonomous vehicles. Meanwhile, compared with the benchmark decision model, the improved reward function model exhibits an 11.33% increase in convergence rate, which fully demonstrating the superiority of the proposed method. Reference | Related Articles | Metrics | Comments（0）

Select

RUL Prediction Model Combined with Transformer ZHENG Hong, LIU Wen, QIU Junjie, YU Jinhao Journal of Applied Sciences    2024, 42 (5): 847-856.   DOI: 10.3969/j.issn.0255-8297.2024.05.011 Abstract （278）      PDF（pc） （711KB）（179）       Knowledge map    Save Remaining useful life (RUL) prediction is crucial for prognostics and health management of large equipment. However, nonlinear characteristics such as high dimensionality, large scale, strong coupling, and time-varying parameters in monitoring data of some devices can lead to low accuracy in RUL prediction. To solve this problem, this paper introduces a neural network model that combines a transformer decoder with a multiscale bi-directional long and short-term memory network. This model improves prediction accuracy of the model by integrating global information through a multi-head attention mechanism. Using aviation engines as the research focus, comparative experiments were conducted employing various models on NASA’s C-MPASS dataset. The results show that the proposed multi-scale bi-directional long and short-term memory network fused with Transformer model (MSBiLSTM-Transformer) outperforms other benchmark models, demonstrating superior performance in both accuracy and root mean square error metrics. Reference | Related Articles | Metrics | Comments（0）

Select

ECG-UNet: a Lightweight Medical Image Segmentation Algorithm Based on U-Shaped Structures PEI Gang, ZHANG Sunjie, ZHANG Jiapeng, PANG Jun Journal of Applied Sciences    2024, 42 (6): 922-933.   DOI: 10.3969/j.issn.0255-8297.2024.06.003 Abstract （711）      PDF（pc） （2125KB）（179）       Knowledge map    Save In recent years, Transformer models have addressed the limitations of deep neural networks in traditional medical image segmentation. However, they still underperform in segmentation at the edges of medical images and suffer from large number of parameters and computational complexity, making them unsuitable for mobile applications. In this paper, we propose a lightweight network called ECG-UNet to mitigate these issues. Firstly, the model uses a strategy combining linear mapping and attention instead of conventional convolution at the bottleneck to reduce the number of network parameters while maintaining performance. Meanwhile, we introduce a lightweight multilayer perceptron module to learn more location information of the image. Secondly, dilated convolutions are applied to expand the respective field. Finally, in exchange for further improvement of the model performance at a relatively small computational cost, a gate attention mechanism is added in the skip connections to enhance the feature propagation in the network. The model is validated on the BUSI and ISIC2018 datasets. The results show that the proposed network structure greatly reduces the computational costs while achieving superior segmentation performance compared to current mainstream algorithms. Reference | Related Articles | Metrics | Comments（0）

Select

Smart Contract Vulnerability Analysis and Improvement Based on Smartcheck FEI Jiajia, ZHAO Xiangfu, CHEN Xiaohan, ZHANG Dengji Journal of Applied Sciences    2024, 42 (6): 1027-1039.   DOI: 10.3969/j.issn.0255-8297.2024.06.011 Abstract （597）      PDF（pc） （1568KB）（176）       Knowledge map    Save Smart contracts on blockchain operate on quantity of digital assets. Once deployed on blockchain, they are difficult to modify. Therefore, the analysis and detection of security vulnerabilities of smart contracts has become an important research topic. Smartcheck is a static analysis tool for Ethereum smart contracts that converts Solidity source code into an XML-based intermediate representation and checks it against XPath patterns. While Smartcheck can analyze most of the vulnerabilities, it has limitations in terms of coverage and accuracy. To address these issues, we developed a new tool, SmartETH, to further improve Smartcheck by analyzing typical vulnerabilities such as timestamp dependency, integer overflow and delegatecall vulnerabilities. The improved Smartcheck is tested on a large dataset and verified by five specific contracts, demonstrating improved accuracy in vulnerability detection. In addition, improvements have reduced false positives and missed positives for many vulnerabilities. Reference | Related Articles | Metrics | Comments（0）

Select

Classroom Expression Classification Model Based on Multitask Learning HE Jiabei, ZHOU Juxiang, GAN Jianhou, WU Di, WEN Xiaoyu Journal of Applied Sciences    2024, 42 (6): 947-961.   DOI: 10.3969/j.issn.0255-8297.2024.06.005 Abstract （631）      PDF（pc） （1155KB）（155）       Knowledge map    Save Facial expression recognition and learning sentiment analysis based on classroom video image understanding have become research hotspots in smart education. However, these applications often face great challenges in real-world scenarios with low-quality image and video acquisition, and serious multi-target occlusion in complex environments. In this paper, a multitask recognition model for classifying student expressions is proposed. Firstly, this study constructs a multitask classroom expression dataset and effectively alleviates the imbalance of class label distribution in the dataset. Secondly, a classroom expression classification model based on multitask learning is proposed. By introducing knowledge distillation and designing a dual-channel fusion mechanism, the model effectively integrates the three tasks of discrete expression recognition, facial action unit detection and valence-arousal estimation. This integration leverages the relationship between multitasks to further enhance the performance of discrete expression classification. Finally, the proposed method is compared with the existing advanced methods across multiple datasets. Results show that the proposed model effectively improves the accuracy of expression classification, and demonstrates superior performance in the multitask recognition of classroom expressions, which provides technical support for multi-dimensional evaluation and analysis of classroom emotions. Reference | Related Articles | Metrics | Comments（0）

Select

NBATMAN-ADV Routing Protocol for Large-Scale Flying Ad Hoc Networks WANG Cong, ZHAO Jihang, WU Xia, MA Wenfeng, TIAN Hui Journal of Applied Sciences    2024, 42 (5): 837-846.   DOI: 10.3969/j.issn.0255-8297.2024.05.010 Abstract （211）      PDF（pc） （3107KB）（145）       Knowledge map    Save Flying ad hoc network is a hot topic in current research, particularly concerning the design of routing mechanisms. The primary challenge lies in managing routing overhead, which can lead to network collapse as the number of UAV nodes increases. To address this issue in large-scale UAVs scenarios, a virtual backbone network is constructed using the unifying connected dominating set algorithm, thereby reducing the number of nodes in route flooding. Next, the NBATMAN-ADV (new better approach to mobile ad-hoc networking-advanced) routing protocol is deployed on the backbone nodes. This protocol evaluates link quality using the received signal strength index and signal-to-noise ratio of the physical layer data, enabling rapid detection of link changes while reducing the routing overhead. Simulation results show that the proposed routing protocol has significantly improved packet delivery rate, end-to-end delay and throughput compared with traditional proactive routing protocols such as optimized link state routing and destination-sequenced distance vector. Experimental results on communication module show that the proposed routing protocol exhibits superior performance in terms of multi-hop delay. Reference | Related Articles | Metrics | Comments（0）

Select

Highway Toll Evasion Patterns Identification Based on RFE-OPTUNA-XGBoost Model MA Feihu, LEI Haoan, SUN Cuiyu, LUO Jiajie Journal of Applied Sciences    2024, 42 (5): 857-870.   DOI: 10.3969/j.issn.0255-8297.2024.05.012 Abstract （183）      PDF（pc） （1156KB）（139）       Knowledge map    Save Driven by economic benefits, highway toll evasion behavior occurs frequently in China. This study utilizes anonymized toll data from a specific region in 2020 to address this issue. Through data mining to analyze the behavior characteristics of the evading vehicles, we propose a toll evasion pattern recognition model based on RFE-OPTUNA-XGBoost. The accuracy of this recognition model reaches 0.945, with average AUC values for different evasion methods as follows: large vehicle misclassification at 0.997, U/J-turn evasion at 0.980, fake green pass at 0.969, and gate crashing at 0.924. The results demonstrate that the RFE-OPTUNA-XGBoost model achieves higher accuracy in toll evasion pattern recognition and higher AUC values for each evasion method. In summary, the proposed model can accurately identify toll evasion patterns, offering significant practical value for highway management departments in conducting inspections and preventing toll evasion. Reference | Related Articles | Metrics | Comments（0）

Select

Encrypted Traffic Classification Based on Attention Temporal Convolutional Network JIN Yanliang, CHEN Yantao, GAO Yuan, ZHOU Jiahao Journal of Applied Sciences    2024, 42 (4): 659-672.   DOI: 10.3969/j.issn.0255-8297.2024.04.008 Abstract （264）      PDF（pc） （609KB）（136）       Knowledge map    Save Aiming at the problem that most current encrypted traffic classification methods ignore the timing characteristics in the traffic and the model efficiency, we propose an efficient classification method based on attention temporal convolutional network (ATCN). This method first embeds content information and timing information into the model to enhance the representation of encrypted traffic. Then it utilizes temporal convolutional network to capture effective features in parallel to increase training speed. Finally, we introduce attention mechanism to establish dynamic feature aggregation to optimize model parameters. Experimental results show the superior performance of our proposed method over the baseline in two classification tasks, achieving accuracy of 99.4% and 99.8%, respectively, while reducing the number of model parameters to a maximum of 15% of the baseline. Finally, a fine-tuning method based on transfer learning is introduced to the ATCN, which provides a novel approach for zero-day traffic processing in traffic classification. Reference | Related Articles | Metrics | Comments（0）