Table of Content

30 November 2024, Volume 42 Issue 6

Previous Issue    Next Issue

Communication Engineering

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

2024, 42(6):  903-911.  doi:10.3969/j.issn.0255-8297.2024.06.001

Asbtract ( 538 )   PDF (2185KB) ( 190 )  

References | Related Articles | Metrics

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.

Normalized Min-sum LDPC Decoding Algorithm Based on Residual Difference Layer

LI Guiyong, WANG Yangyang, LIANG Zhiyong

2024, 42(6):  912-921.  doi:10.3969/j.issn.0255-8297.2024.06.002

Asbtract ( 534 )   PDF (758KB) ( 73 )  

References | Related Articles | Metrics

In order to further narrow the gap between min-sum (MS) algorithm and belief propagation (BP) algorithm, and to improve the decoding performance of normalized minsum (NMS) algorithm, an improved normalized minimum sum LDPC decoding algorithm based on residual difference layer is proposed. Firstly, the overestimation problem of MS algorithm is quantitatively analyzed. BP algorithm and MS algorithm are used to test the ratio characteristics of node LLR messages, and the corresponding normalization factors are calculated. To reduce decoding complexity, a weighted average processing is adopted according to the variation in the optimal normalization factor. Additionally, to reduce the average number of iterations and accelerate decoding convergence, the proposed algorithm uses the residual characteristics of the check node information to prioritize updates in layers with larger residual values. The layers are dynamically rearranged between iterations. Simulation results show that the proposed RB_LINMS algorithm achieves a performance gain of approximately 0.26 dB in decoding, compared with the traditional NMS algorithm at the bit error rate 10^-5, and reducer the average number of iterations by up to 33.20%. Therefore, with a slight increase in complexity, it offers faster convergence and improved decoding performance.

Signal and Information Processing

ECG-UNet: a Lightweight Medical Image Segmentation Algorithm Based on U-Shaped Structures

PEI Gang, ZHANG Sunjie, ZHANG Jiapeng, PANG Jun

2024, 42(6):  922-933.  doi:10.3969/j.issn.0255-8297.2024.06.003

Asbtract ( 616 )   PDF (2125KB) ( 152 )  

References | Related Articles | Metrics

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.

A Low Light Image Enhancement Method Based on CRTNet

JIANG Zetao, HUANG Jingfan, ZHU Wencai, HUANG Qinyang, JIN Xin

2024, 42(6):  934-946.  doi:10.3969/j.issn.0255-8297.2024.06.004

Asbtract ( 488 )   PDF (4085KB) ( 209 )  

References | Related Articles | Metrics

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.

Classroom Expression Classification Model Based on Multitask Learning

HE Jiabei, ZHOU Juxiang, GAN Jianhou, WU Di, WEN Xiaoyu

2024, 42(6):  947-961.  doi:10.3969/j.issn.0255-8297.2024.06.005

Asbtract ( 555 )   PDF (1155KB) ( 132 )  

References | Related Articles | Metrics

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.

Point Cloud Registration Method Based on Principal Component Eigenvectors

ZHAO Fuqun, HUANG He, GENG Guohua

2024, 42(6):  962-976.  doi:10.3969/j.issn.0255-8297.2024.06.006

Asbtract ( 523 )   PDF (12494KB) ( 73 )  

References | Related Articles | Metrics

To address the issues of low accuracy and long time consumption of the existing point cloud registration algorithms for cluttered point clouds, a point cloud registration method based on principal component eigenvectors is proposed. Firstly, feature point set is extracted by describing the curvature change of the point cloud, and the center of gravity method is applied to align the center of gravity of the reference point cloud with that of the feature point set, achieving an initial rough registration. Then, during the iterative closest point (ICP) algorithm, principal component analysis (PCA) is used to select the first three principal component feature vectors and perform corresponding matching through rigid body transformation. Lastly, the Euclidean distance is used to find the nearest points for fine registration. The proposed method was validated using both public point cloud and cultural relic point cloud. Experimental results show that the registration accuracy of the proposed method is improved by approximately 12% on average, while the registration time is reduced by about 10% on average. These results indicate that the proposed method based on principal component eigenvectors is an effective approach for point cloud registration.

Retrieval of Lake Surface Temperature in Plateau Based on Landsat-8 Images

HANG Yuanfang, ZHANG Penglin

2024, 42(6):  977-987.  doi:10.3969/j.issn.0255-8297.2024.06.007

Asbtract ( 586 )   PDF (645KB) ( 76 )  

References | Related Articles | Metrics

At present, there are few studies on water surface temperature retrieval in the plateau environments. This paper combined Landsat-8 image and radiative transfer equation method to invert the water surface temperature in Namco Lake and verify its effectiveness. First, the surface-specific emissivity of the study area is calculated. Second, the radiation brightness of the blackbody at the same temperature is determined. Finally, the Planck function is used to calculate the water surface temperature and MODIS surface temperature products is used to validate the inversion results. Experimental results show that the minimum absolute error of lake surface temperature retrieval is 0.449 ℃, with a maximum error of 1.685 ℃. The minimum root mean square error ranges from 1.269 ℃ to 1.781 ℃. The inversion results are close to the average daily temperature results from MODIS products. The variation characteristics of lake surface temperature in summer are basically consistent with the temperature of Namco Lake. This method provides a reference for future research on water surface temperature retrieval in plateau lakes.

A Calculation Method for Tilt Angle of Power Tower Based on Laser Point Cloud

HUANG Kewen, MENG Yanxi, YU Haotian, JIA Tao

2024, 42(6):  988-999.  doi:10.3969/j.issn.0255-8297.2024.06.008

Asbtract ( 638 )   PDF (10902KB) ( 61 )  

References | Related Articles | Metrics

This paper addresses the challenge of accurately measuring the tilt angle of power towers, which is limited by low computational efficiency, tedious process, and high difficulty in conventional manual inspections. We propose a novel method for calculating the tilt angle using point cloud data obtained from unmanned aircraft. Our method starts by acquiring the point cloud data of power tower by unmanned aircraft, followed by the extraction of their major structure, considering both statistical characteristics and structural components. The centerline of the power tower is then computed and used in conjunction with the plumb line of the ground to determine the tilt angle. To verify the effectiveness of our method, experiments are conducted by applying it to four power towers in different locations. The accuracy of the calculated tilt angles is analyzed through numerous simulations and influential factor analysis. The results suggest that our method achieves a high accuracy of 0.017 4° on average, demonstrating the effectiveness and reliability for use in conventional tower inspection.

Computer Science and Applications

Unbalanced Multiclassification Study Based on Mixed Sampling and SE_ResNet_SVM

JIAO Guie, WENG Tongtong, ZHANG Wenjun

2024, 42(6):  1000-1015.  doi:10.3969/j.issn.0255-8297.2024.06.009

Asbtract ( 560 )   PDF (837KB) ( 71 )  

References | Related Articles | Metrics

A network model SNSMRS (SMOTEENN-mixed residual networks-SVM network) based on hybrid sampling, squeeze and excitation (SE) module, improved deep residual network and support vector machines (SVM) is proposed to address the problem of uneven class distribution of unbalanced data sets in traditional structured multiclassification algorithms, which leads to increased classification difficulty. Firstly, the data distribution is improved by synthesizing minority oversampling and editing nearest neighbors technique. Then the features are extracted by combining SE module and a deep residual network, improved with batch normalization and group normalization. Finally, the network model uses support vector machine (SVM) to output the classification results. The SE module enhances the model’s feature differentiation ability and robustness. The improvements to the ResNet, through fusion normalization, mitigate issues such as gradient vanishing and accuracy degradation, and ensure stability and accuracy regardless of batch_size. Additionally, SVM enhances the classification accuracy by effectively utilizing feature vectors in space to classify and extract features. Comparison and ablation experiments are conducted on seven unbalanced public datasets of various sizes and domains. The experimental results show that the proposed model, SNSMRS, not only outperforms other deep learning models, but also increases the values of Macro-F1 and G-mean by approximately 3% and 4%, respectively, compared with the original ResNet. Macro-F1 and G-mean values of SNSMRS exceed 95% on four of the datasets, demonstrating its superior performance.

Emotion Recognition of EEG Using Subdomain Adaptation and Spatial-Temporal Learning

TANG Yiheng, WANG Yongxiong, WANG Zhe, ZHANG Xiaoli

2024, 42(6):  1016-1026.  doi:10.3969/j.issn.0255-8297.2024.06.010

Asbtract ( 608 )   PDF (1384KB) ( 131 )  

References | Related Articles | Metrics

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.

Smart Contract Vulnerability Analysis and Improvement Based on Smartcheck

FEI Jiajia, ZHAO Xiangfu, CHEN Xiaohan, ZHANG Dengji

2024, 42(6):  1027-1039.  doi:10.3969/j.issn.0255-8297.2024.06.011

Asbtract ( 528 )   PDF (1568KB) ( 128 )  

References | Related Articles | Metrics

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.

A Multimodal Knowledge Graph Entity Alignment Method

LIU Wei, XU Hui, LI Weimin

2024, 42(6):  1040-1051.  doi:10.3969/j.issn.0255-8297.2024.06.012

Asbtract ( 608 )   PDF (1304KB) ( 156 )  

References | Related Articles | Metrics

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.

Non-isometric Histogram Publishing Algorithm Based on Differential Privacy

SHAN Liyang, CHEN Xuebin, GUO Rumin

2024, 42(6):  1052-1063.  doi:10.3969/j.issn.0255-8297.2024.06.013

Asbtract ( 511 )   PDF (724KB) ( 74 )  

References | Related Articles | Metrics

To address the histogram privacy leakage and the challenge of determining the number of groups, a non-equidistant histogram data publishing algorithm based on differential privacy (DP) is proposed. Firstly, an improved quantified comprehensive evaluation index is introduced, which quantifies the criterion of histogram grouping into a specific calculation formula to determine the optimal number of histogram groups. Next, the empirical distribution function is used to design a privacy budget allocation scheme, and the grouping boundaries are calculated to construct the non-equidistant histogram. The dataset is then divided according to the non-equidistant boundaries, and the frequencies are counted, with noise added to satisfy the differential privacy requirements. The non-equidistant histogram is subsequently published. Experimental results show that the optimal calculation of the number of groups and the implementation of non-equidistance can ensure the accuracy and privacy of the published data of the histogram, while preserving the distribution characteristics of the histogram. The mean square error of the proposed algorithm is reduced by 99% compared with similar accurate histogram publication (AHP) algorithms.

User Identification Method Using Proximity and Content Features

LU Jing, YOU Chenlu, GAI Qikai, LIU Cong

2024, 42(6):  1064-1077.  doi:10.3969/j.issn.0255-8297.2024.06.014

Asbtract ( 498 )   PDF (734KB) ( 21 )  

References | Related Articles | Metrics

Social networks restrict access to user topology, which greatly reduces the accuracy of identification methods using structure features. We present proximity and content based User Identification based on XGboost, a semi-supervised network model that integrates attribute, structural and content features to transform the cross-social network user identification problem into a binary classification task. To tackle the challenge of incomplete topology information and insufficient seed users, a method for extracting explicit and implicit friends is proposed. Friend networks are fused according to explicit friends, implicit friends and other friends in the friend network of the user pair to be matched. The user’s importance is combined, so as to improve empirical probability of second order proximity of LINE algorithm and obtain the structure feature. We then extract time sequence features, keyword overlapping features, and followee tag feature as the content features. Finally, these features are fused to complete user identification. Experiments on real datasets show the effectiveness of this method.

Electronic Engineering

Research on Fatigue Failure Model of IGBT Power Module Based on Steady-State Collector-Emitter Saturation Voltage

LU Jing, LI You

2024, 42(6):  1078-1088.  doi:10.3969/j.issn.0255-8297.2024.06.015

Asbtract ( 514 )   PDF (796KB) ( 61 )  

References | Related Articles | Metrics

With the increasing application of insulated gate bipolar transistor (IGBT) power module in both civil and military fields, the issue of fatigue aging under electrothermal stress has gained critical importance. Based on the theory of semiconductor physics and device reliability, this paper investigates package failures of IGBT power modules, and analyzes the mechanism and performance characteristics of solder layer and bonding line failures. By examining the IGBT conduction model, we propose a novel fatigue aging model based on the steady-state collector-emitter saturation voltage, enabling comprehensive characterization of packaging problems such as solder layer fatigue and wire fatigue. To validate the model, an electro-thermal aging test platform is built. Experimental results from cyclic aging tests verifies that the proposed fatigue aging model can accurately evaluate the fatigue aging degree of IGBT power module packages.