最新刊期
卷 23 ， 期 10 ， 2024
- “在工业软件领域，专家提出资源共享技术，构建知识图谱，利用人工智能提升资源利用率和应用水平，为国产软件产业化奠定基础。”摘要：As an important productivity tool， industrial software serves as the foundation supporting the development and innovation of large and complex equipment manufacturing enterprises. Its application level determines the enterprise product value and costs， which are crucial for enhancing the core competitiveness of large complex equipment manufacturing enterprises. Based on the current research and development status， this paper proposed industrial software resource sharing technology， significantly increasing the resource utilization and application level of industrial software. They have constructed an industrial software knowledge graph， facilitating the positive accumulation of industrial software knowledge. By employing artificial intelligence technology， they analyzed the characteristics of industrial software and its knowledge， enabling more precise scheduling of industrial software and intelligent recommendation of knowledge for different usage scenarios， user interests， and behaviors. This enhanced the intelligence and convenience of industrial software resource sharing services. Furthermore， based on industrial software resource sharing technology， a data-driven industrial software application effectiveness evaluation model and methods were proposed， laying the groundwork for the maturation and industrialization of domestic industrial software. Through a case study of establishing an industrial software resource pool within a certain aerospace research institute， the advantages of industrial software resource sharing and application evaluation technology， as well as its demonstrative role in large complex equipment manufacturing enterprises were analyzed. Lastly， the development direction of industrial software resource sharing technology was discussed in conjunction with the construction of the domestic industrial software ecosystem.关键词：industrial software;industrial software sharing;application effectiveness evaluation;industrial software ecosystem construction80|208|0更新时间：2024-10-14
- “在国家数值风洞工程支持下，FlowStar软件在航空航天等领域取得显著进展，为HPC与CFD融合发展提供新思路。”摘要：Based on the National Numerical Wind Tunnel Engineering Project， NNW-FlowStar is an industrial general-purpose unstructured computational fluid dynamics software. It incorporates a diverse range of numerical models and intricate functional modules，and it has been widely applied in the aerospace and other relevant fields. With the advancement of high-performance computer architectures， optimizing large-scale industrial-grade CFD software such as FlowStar on HPC systems has become increasingly challenging on HPC systems.Leveraging typical HPC platform，this work conducts in-depth testing and analysis through the approach of the computation， communication， and memory access characteristics. It helps researchers to understand the performance characteristics of complex CFD software on HPC systems deeply and provids recommendations and references for future futher parallel computing and performance optimization of FlowStar. This study serves as a reference for parallel computing and performance optimization in other non-structured CFD software and can also contribute to the integration and development of HPC and CFD.关键词：CFD parallel computing;FlowStar;performance testing;performance analysis56|181|0更新时间：2024-10-14
- “在先进制造研发领域，专家开发了多物理场耦合仿真软件，提升了核心技术产业的设计水平和制造效率。”摘要：With the in-depth development of simulation technology， multi-physics field coupling calculation has become an important trend in engineering research and development. In order to solve the problems of data dispersion， inefficient process convergence and difficult sharing of knowledge and experience， and low R&D efficiency in the simulation system， a set of multi-physics field coupling simulation software for advanced manufacturing R&D scenarios is developed， the system of multi-physics field simulation software is improved in depth， and a domestically-owned multi-physics field simulation software system is established， and low-frequency magnetic and fluid simulation software is developed to complete the heat-fluid-solid， electro-magnetic-heat ， solver coupling of electro-thermal coupling simulation， and integrated into the multiphysics field simulation platform with unified data transmission， convenient design modeling， fast computational efficiency， and high accuracy of coupling results. Example verification results show that the application of the software can be obtained from the design input， geometric analysis， mesh division， multi-physics field setup， multi-field coupling solution， and result output within the effective link to obtain reasonable and high precision simulation results， which improves the overall design level and manufacturing efficiency of aerospace， new energy vehicles， semiconductors， computers， and other core technology industries.关键词：multi-physics field;mesh sectioning;coupling;electro-thermal force;electromagnetic heat;simulation software177|463|0更新时间：2024-10-14
- “在智能无人机器人领域，专家提出了一种高阶滑模控制方案，有效解决了反作用力轮无人自行车的独立平衡问题，为精确稳定运动提供解决方案。”摘要：A coordinated motion control scheme is crucial for the precise and stable motion of intelligent unmanned robots. A high-order sliding mode control scheme combining fixed time convergence and adaptive disturbance compensation is proposed for the independent balancing problem of reaction wheel unmanned bicycles. This scheme can be widely applied to problems such as actuators that consider amplitude and rate constraints. Firstly， considering the overall design， a fixed time fast convergence sliding mode controller is adopted for the power drive of the reaction wheel unmanned bicycle. At the same time， a high-order sliding mode disturbance observer is designed to compensate for the unmodeled parameter terms and external disturbances. Then， the stability proof of the closed-loop control system is given based on Lyapunov functions. In practical scenarios， verification experiments were conducted， and the proposed scheme showed a performance improvement of 5% to 30% compared to other sliding mode control methods， especially in terms of dynamic driving error RMSE reaching 0.003 1 rad， indicating that the controller can be stable and effective during driving.关键词：reaction wheel unmanned bicycle;independent balance;disturbance observer;higher-order sliding mode control33|186|0更新时间：2024-10-14
-
Research on the Full Lifecycle Data Integration of Nuclear Engineering Based on Data Consistency AI导读
“在核工程领域，专家提出以产品和业务主数据为核心的全周期数据集成框架，助力企业数字化转型。”摘要：The comprehensive implementation of the strategic task of "digital nuclear industry" poses higher requirements for data integration throughout the full lifecycle of nuclear engineering. Data consistency is an important foundation for full cycle data integration. Based on the extensive practice and development trend of data integration， the classification concepts of nuclear engineering resource master data， product master data， management master data， and business master data are clarified， and an application architecture for data integration based on product master data consistency is proposed. Compared to product master data， data integration with business master data as the core has the characteristics of wider scope and finer granularity. This further elaborates on the overall framework of nuclear engineering full cycle data integration with product and business master data as the core， in order to help nuclear engineering enterprises achieve a wider range and deeper application of digital transformation.关键词：data consistency;nuclear engineering;full lifecycle data integration;master data19|119|0更新时间：2024-10-14 - 摘要：A power engineering project audit method based on event extraction and verification warning is proposed to address the difficulties in integrating audit management and digital operations caused by the wide scope of business， multiple audit processes， and complex event correlations in power grid engineering projects. This method first constructs an audit business process based on an event framework to reduce the workload and business expansion pressure of modeling personnel；Then， using template matching combined with Transformer， the audit events recognized from fixed format text are used as remote learning knowledge to generate trigger word templates with clear prompt knowledge， and knowledge is supplemented as needed according to different project topics， thereby improving the accuracy of event extraction from non fixed format text； Finally， the identified audit events will be matched to different audit stages and event frameworks to support the verification and early warning of relevant audit events in accordance with project progress. The verification experiment shows that in low manually annotated text， the accuracy of event extraction using this audit method is not less than 93%， and the extracted events can be accurately embedded into the audit business process， supporting abnormal verification and early warning of related audit events.关键词：audit business;event framework;event extraction;Transformer;template generation;verification warning20|131|0更新时间：2024-10-14
- “在半导体缺陷检测领域，研究人员提出了一种基于改进自编码器和深度特征提取器的晶圆表面缺陷检测模型，有效提升了缺陷检测精度。”摘要：In today's semiconductor defect detection field， it always faces the problem of insufficient defect samples and diversified defect samples， in order to solve the problem effectively， a wafer surface defect detection model based on improved self-encoder and deep feature extractor is proposed by using the DFR model as the basic framework， which achieves a better feature extraction by using the pre-trained VGG19 model as a feature extractor， and subsequently image reconstruction using improved self-encoder to learn the normal features of the image. The anomaly scores are obtained by comparing the global differences between the input and generated images for defect detection， and the results show that for the homemade wafer dataset， the average AUC improves by 0.8 percentage points compared to the baseline model， and the accuracy of defect detection reaches 0.997； for the MVTec AD dataset， the average AUC improves by 2.5 percentage points compared to the baseline model， and the accuracy of defect detection reaches 0.963.关键词：defect detection;feature extraction;autoencoder;AFF attention mechanism34|115|0更新时间：2024-10-14
- “在新型工业化背景下，湖北省工业软件生态体系研究为提升产业链稳定性和安全性提供借鉴。”摘要：Industrial software has been a crucial support in the high-quality development of manufacturing industry. In the process of new industrialization， to expand the ecosystem of industrial software is of significant importance to accelerate the development of modern industrial system. Based on the analysis to the strategic positioning of new industrialization， this paper summarizes the trends of both the international and domestic software industry. The developing status of industrial software in Hubei Province and the challenges confronting are introduced. Under such circumstances， countermeasures towards the high-quality development of industrial software in Hubei Province are put forwarded through using the experience of the advanced experience at home and abroad. This research caters for the national strategic goals and aims at raising the stability， security and competences in key industries.This research caters for the national strategic goals and aims at raising the stability， security and competences in key industries. All these can be used as references for the development of industrial software and the effectiveness of the whole industrial and supply chain in Hubei Province.关键词：industrial software;new industrialization;high-quality development;industrial chain43|94|0更新时间：2024-10-14
- “电子科技大学软件学院在EDA工业软件专业建设方面取得初步成果，融合多学科内容是关键。”摘要：At present， the design and development of electronic automation design （EDA） industrial software is one of the bottleneck technologies in China's industrial development， and there is an urgent need to cultivate relevant professional talents.The construction of EDA industrial software majors in China has just begun， and many aspects are still in the exploratory stage. In response to this issue， the School of Software at the University of Electronic Science and Technology of China has established a faculty team from scratch， developed teaching syllabi and lesson plans， selected suitable textbooks， built experimental platforms， trained teachers， strengthened cooperation with domestic EDA companies， and achieved preliminary results in the construction of EDA direction for characteristic software engineering majors. Summarizing the experience， difficulties， ideas， experiences， and lessons learned in the construction process， it is believed that the integration of software engineering， microelectronics， computer science， applied mathematics， and physics is the key.关键词：electronic design automation;major construction;industrial software16|112|0更新时间：2024-10-14
- “在海上交通安全管理领域，专家提出了一种结合卷积神经网络、改进长短时记忆网络与注意力机制的船舶轨迹预测模型，提高了模型精度。”摘要：With the acceleration of economic globalization and the continuous expansion of international trade， the maritime transportation industry is developing rapidly. In ports with high traffic density and complex conditions， traffic safety management is facing enormous challenges. Ship collision is one of the frequent types of accidents at sea， and accurate ship prediction is extremely important for maritime traffic management and ensuring the safety of ship navigation. The commonly used method for predicting ship trajectories is the Long Short Term Memory Network， but it has a large number of gate control weight parameters， a complex structure， and insufficient exploration of spatial and temporal features. A ship trajectory prediction model combining convolutional neural network， improved long short-term memory network， and attention mechanism is proposed to address the above issues. This model reduces structural complexity， improves training speed and generalization performance through an improved long short-term memory network； At the same time， convolutional neural networks are introduced to fully explore the spatial and temporal features of trajectory data， and different weights are assigned to different features through attention mechanisms to filter out useless feature information and improve model accuracy. The experimental results on real datasets show that the proposed model has improved accuracy in predicting latitude， longitude， heading， and speed compared to mainstream control models.关键词：track prediction;data preprocessing;deep learning;natural language processing;AIS data103|202|0更新时间：2024-10-14
- “在无人机勘探领域，ICOA算法通过改进浣熊算法，提高了全局搜索能力，为丘陵地区无人机路径规划提供有效解决方案。”摘要：To solve the problem of insufficient global searching ability of coati optimization algorithm （COA）， this paper proposes an improved coati optimization algorithm （ICOA）， which initializes the population through SPM chaotic mapping， introduces a comprehensive position update strategy of Levy flight and lens imaging reverse learning， and improves the ability of jumping out of local optimal solution and global searching. Benchmark function test results show that ICOA has better convergence speed and convergence accuracy than COA， sparrow search algorithm （SSA） and arithmetic search algorithm （AOA）. Aiming at the path planning problems of various terrain， threats and constraints of 3D UAVs， the simulation environment of hilly landform is constructed to simulate the scene of UAVs performing exploration tasks of ancient buildings. Meanwhile， ICOA， COA， SSA and AOA are used for path planning. The simulation results show that the fitness value of the path planned by ICOA algorithm is the best， the path distance is shorter and the pitch Angle is smaller， which further verifies the effectiveness of ICOA and the feasibility of its application in UAV exploration missions in hilly areas.关键词：improved coati algorithm;path planning;SPM chaotic mapping;Lévy flight;lens imaging reverse learning18|122|0更新时间：2024-10-14
- “在股票交易领域，专家将邻近森林算法与自动化交易订单结合，建立了量化交易系统，平均年回报率达20.09%。”摘要：By studying the application of the neighboring forest algorithm in stock trading， a trading signal classification model is obtained by combining the neighboring forest classification model with automated trading orders. Firstly， the classification model is combined with the index moving average strategy to obtain a dual layer filtering model for trading signals. Then， the pricing strategy for automated orders， stock selection， and trading strategy for determining shares are determined through the average amplitude indicator and Kelly formula； Finally， the double-layer filtering model is combined with pricing and trading strategies to obtain a quantitative trading system based on the proximity forest algorithm. Simulation experiments show that the average annual return rate of the system reaches 20.09%. Compared to the strategy relying solely on the exponential moving average， the neighboring forest algorithm has a more significant advantage in profitability.关键词：proximity forest;auto trading order;risk management;trading system25|98|0更新时间：2024-10-14
- “基因组框架填充技术取得新进展，专家提出解决方案，为提高基因组完整性提供新思路。”摘要：The development of gene sequencing technology has continuously improved the speed of obtaining genome sequences， but currently widely used sequencing technologies often produce genome frameworks with partial gene deletions. The use of genome fragment filling technology can effectively improve the integrity of genome frameworks and reduce sequencing costs. In previous studies， missing genes could be inserted between any two genes in an incomplete sequence； In the later stage， there are limitations on the selection of insertion positions for missing genes in genome frameworks provided in the form of contigs and block matching. This type of fragment filling problem is more general. To this end， research was conducted on the problem of single-sided genome frame filling with restricted insertion positions， and the research progress of this problem was analyzed. The core ideas， processes， advantages and disadvantages of existing FPT algorithms and approximation algorithms were discussed in detail. During the research process， issues such as redundant common adjacency and low approximation performance were identified， and corresponding solutions were proposed. Future research directions and challenges were analyzed.关键词：genome;scaffold filling;contig;block-matching;FPT algorithms;approximation algorithm13|46|0更新时间：2024-10-14
- “在优化算法领域，研究者提出了一种融合黏菌算法与切线飞行的克隆选择算法，有效提升了算法的收敛速度和求解精度。”摘要：In view of the fact that the traditional clone selection algorithm may have slow convergence speed and easily fall into the local optimal problem， a clone selection algorithm that combines the slime mold algorithm and tangent flight is proposed. Firstly， incorporate the position updating strategy of the slime mold algorithm to mutate the cloned population， enhancing the global search capability of the clone selection algorithm during the iteration process. Secondly， a curve convergence strategy is proposed to adjust the search ability of the algorithm， thereby improving the convergence speed of the algorithm. Finally， the Tangent flight strategy is utilized to enhance the algorithm's ability to escape from local optima. The proposed strategy was validated by conducting tests on 14 test functions， compared with other intelligent algorithms and improved CSA algorithm， and conducted Wilcoxon rank-sum test. The experiments show that the convergence speed and solution accuracy of the algorithm， were improved and verifies the good optimization ability of the improved algorithm.关键词：clone selection algorithm;intelligent algorithm;slime mold algorithm;convergence strategy;tangent flight14|87|0更新时间：2024-10-14
-
Research on Cloud Computing Task Scheduling Based on Mixed Strategy Whale Optimization Algorithm AI导读
“在云计算任务调度领域，专家提出了基于混合策略鲸鱼优化算法的调度方法，有效降低了任务执行时间和成本，提升了系统负载均衡性。”摘要：A cloud computing task scheduling method based on the Hybrid Strategy Whale Optimization Algorithm （MSWOA） is proposed to address issues such as long task execution time， high system execution costs， and imbalanced system loads in the process of cloud computing task scheduling. Firstly， use Tent chaotic mapping to initialize the whale population to enhance population diversity and make the distribution of whale individuals more uniform； Then， an adaptive probability threshold was proposed to balance the global search capability and local development capability of the algorithm， and the Levy flight strategy was introduced in the random search stage of the algorithm to expand the search space and search capability of the algorithm； Finally， a multi-objective fitness function was designed for the task scheduling process， and an algorithm was used to solve the multi-objective task scheduling problem in cloud computing. The simulation experiment of MSWOA was conducted using CloudSim cloud computing simulation software， and the results of comparing MSWOA with NOA， ZOA， OAWOA， and TSWOA algorithms showed that compared with other algorithms， MSWOA achieved better performance at different task scales. It not only reduced the maximum completion time and system execution cost of tasks， but also improved the average load rate of the system， which has significant advantages in multi-objective task scheduling in cloud computing.关键词：cloud computing;task scheduling;whale optimization algorithm;multi-objective optimization;Lévy flight60|182|0更新时间：2024-10-14 - “在水下超声无损检测领域，提出了一种基于EWT和FastICA的超声回波信号降噪方法，有效提高了缺陷检测精度。”摘要：Aiming at the problem that underwater ultrasonic non-destructive testing is susceptible to noise signal interference， resulting in low defect detection accuracy， an underwater ultrasonic echo signal detection method based on empirical wavelet transform combined with fast independent component analysis is proposed. First， the EWT algorithm is used to decompose the ultrasonic echo signal to obtain the intrinsic mode function（IMF） of different scales； then the fuzzy entropy value of each modal component is calculated by the fuzzy entropy algorithm， and the required modal components are screened out； finally the FastICA operation is performed to reconstruct the screened IMF， and the blind source separation of the ultrasonic echo signal and the noise signal is realized， and a pure ultrasonic echo signal is obtained in the end. The simulation and experimental examples show that the noise reduction method combined with EWT-FastICA improves the mode mixing， endpoint effect， over-envelope and under-envelope phenomena that occur in traditional empirical mode decomposition， the noise reduction process is faster and more accurate， and can be applied to similar underwater ultrasonic detection signal noise reduction processing.关键词：underwater ultrasonic non-destructive testing;ultrasonic signals;fuzzy entropy value;EWT;FastICA11|64|0更新时间：2024-10-14
-
Model of Rumor Propagation in Online Social Networks Considering the Identity-Skepticism Mechanism AI导读
“在在线社交网络谣言传播研究中，专家建立了修正的SIR模型，为控制谣言传播提供新思路。”摘要：Rumor spreading in online social networks is influenced by a variety of factors. Considering that netizens have agreeable and skeptical attitudes towards rumors， some of them believe the rumors and spread them， another part does not believe the rumors and is immune to them， and another part will keep the state of unknown or neutral and so on. Based on the traditional SIR rumor propagation model， the unknowns' identity-skepticism mechanism is introduced， the identity-skepticism probability is considered to be related to the number of rumor retweets and the correlation coefficient， and the rumor spreader's forgetting rate is related to the timeliness of the information， and the modified SIR model is established， and the mean-field equations of the modified SIR rumor propagation model are firstly derived for the uniform network and the inhomogeneous network， and then the Monte Carlo method is used to simulate the rumor The rumor propagation process was simulated using the Monte Carlo method. The simulation results show that the decrease of information timeliness， the increase of forgetting rate and the probability of questioning the rumor will lead to the decrease of the peak density of spreaders and the scale of rumor spreading； the more times the rumor is forwarded， as well as the higher the relevance of the rumor to the people， and the wider the scope of spreading， the higher the peak density of spreading nodes.关键词：online social networks;rumor spreading;identity-questioning mechanism;informationtimeliness;Monte Carlo simulation29|93|0更新时间：2024-10-14 - “在推荐系统领域，最新研究提出了基于图神经网络的UEKR算法，通过挖掘潜在偏好图，有效提升了推荐性能。”摘要：Graph recognition plays an increasingly important role in recommendation systems， and the latest technological trend is to develop end-to-end recommendation models based on graph neural networks. However， existing GNN based models often fail to fully explore the information in the knowledge graph， simply connecting users to entities in the knowledge graph through projects， without clearly modeling the relationships between users and entities. To this end， a recommendation algorithm UEKR based on graph neural networks is proposed for mining latent preference maps. It dynamically extracts entities of interest to users from collaborative knowledge graphs， models the relationship between users and entities， and constructs a user entity relationship graph to enrich user representation and enhance recommendation performance. The experimental results on three benchmark datasets showed that UEKR improved AUC indicators by 0.75% to 3.65% and F1 indicators by 0.70% to 1.75% compared to the control model.关键词：recommender system;knowledge graph;graph neural network;deep learning48|54|0更新时间：2024-10-14
- “区块链技术助力企业征信数据管理，构建跨区域公共资源交易链，实现全生命周期管理，满足实际业务需求。”摘要：In contemporary corporate credit reporting， information is scattered across governmental and business sectors， resulting in a information silo phenomenon. The extent and richness of data openness are markedly insufficient. Moreover， difficulties arise in evidence preservation for data usage authorization and in random inspections. Relying on a universally interconnected platform with credibility leads to concerns of single-point failures and information leaks. In response， we have integrated blockchain technology into corporate credit data management services. We propose a cross-regional public resource transaction chain tailored for credit reporting service demands. This chain can accommodate the necessities of cross-regional public resource transactions and ensures node decentralization， diversified consensus， and encrypted communication. Based on the underlying public resource transaction chain， we have established a trustworthy smart contract management system， encompassing contract code auditing， automated lifecycle management， and controllable adjudication， thereby achieving flexible management of public resource transaction data. Ultimately， we designed a data management mechanism in line with credit management service standards using smart contracts， accomplishing full lifecycle management of corporate credit data. Through empirical testing， we have verified the system’s performance， confirming its suitability for real- world credit reporting scenarios.关键词：blockchain;public resource trading;data management;credit reporting service18|62|0更新时间：2024-10-14
- “在信息技术领域，专家提出了基于SM9的API KEY强安全应用方案，有效提升了API KEY的安全性和可靠性。”摘要：With the rapid development of information technology， the importance of APIs in various applications and services is becoming increasingly prominent. However， traditional API keys have many security risks in applications. Therefore， a strong security application scheme for API KEY based on SM9 is proposed. Firstly， use SM9 to sign and verify the data； Then introduce a trusted third-party to manage user public and private key pairs， and combine them with random status codes to improve the security and reliability of API KEY； Finally， compare the security of API KEY applications with Baidu Open Platform and FOFA Network Space Surveying Platform. The experiment shows that the API KEY strong security application scheme based on SM9 has advantages such as data integrity， source reliability， resistance to leakage attacks， resistance to replay attacks， resistance to CSRF attacks， and resistance to brute force attacks.关键词：API KEY;SM9;strong security;trusted third party;public and private key pairs11|63|0更新时间：2024-10-14
- “在网络安全领域，专家基于人际欺骗理论，探究了社交情境对鱼叉式网络钓鱼易感性的影响，为降低网络钓鱼易感性提供了参考。”摘要：In order to reduce the serious harm of harpoon phishing， this paper explores the susceptibility of harpoon phishing in Internet social situations through the interaction of intentional situation induction and unintentional situation involvement. Based on the theory of interpersonal deception， a configuration model is proposed to investigate the influence of social context on susceptibility to spear phishing. The model focuses on the interaction between intentional situational induction by anglers and unintentional situational involvement by anglers. Necessary condition analysis and qualitative comparative analysis are used to empirically test the research hypothesis. Empirical analysis has identified two types of susceptibility configurations for spear phishing， namely high information embedding induction by anglers， high preference matching induction， and high social information involvement configuration by anglers， as well as high information embedding induction by anglers， high trust induction， and high thematic involvement configuration by anglers. The research results extended the susceptibility of spear phishing from within the organization to outside the organization from a social context perspective， and combined with interpersonal deception from a configuration perspective， extended the influence of context on susceptibility to spear phishing from a one-way perspective to a two-way interaction of intentional context induction and unintentional context involvement. This promoted the study of context that affects susceptibility to spear phishing and provided some reference for reducing susceptibility to spear phishing.关键词：spear phishing;interpersonal deception theory;intentional context induction;involving unintentional context;configuration effect9|196|0更新时间：2024-10-14
- “在移动众包网络领域，专家提出了SIR-M恶意代码传播模型，通过众包机制显著减缓了恶意代码传播速度，降低了网络风险。”摘要：Based on the inherent characteristics of mobile crowdsourcing networks （MCN）， dynamic incentives are introduced to promote the activity analysis of mobile users （MU） in the network. The dynamic behavior of malicious code propagation in MCN is studied， and a new malicious code propagation model SIR-M is proposed， where M nodes represent crowdsourcing nodes that handle node tasks. Firstly， considering the initiative of the newly infected node， the infected node can seek MU isolation and immune enhancement for this node through the crowdsourcing mechanism of MCN. Then， the effectiveness of the model was verified through stability analysis and numerical simulation， and compared with the SIR model to analyze the impact of crowdsourcing mechanism on the system. The results indicate that the crowdsourcing mechanism of mobile crowdsourcing networks significantly slows down the propagation speed of malicious code and reduces the risk of large-scale proliferation of malicious code in the network.关键词：propagation model;dynamic incentives;MCN;stability analysis;malware26|80|0更新时间：2024-10-14
- “在可见光通信领域，研究者提出了一种64QAM星座方案，有效提升了传输速率和抗噪声能力。”摘要：In order to improve the transmission rate of visible light communication systems and overcome channel correlation， a superimposed 64 order orthogonal amplitude modulation constellation scheme is proposed in a 2 × 2 multi input multi output visible light communication system. Firstly， generate two geometrically integer 8QAM signals at the transmitting end and send them through two light-emitting diodes； Then， a square 64QAM constellation signal is obtained by linearly superimposing the free space transmission to the receiving end. Experiments have shown that the proposed scheme has a lower bit error rate， a wider range of peak to peak voltage operation， and a larger minimum Euclidean distance and lower peak to average power ratio compared to existing schemes. Therefore， it has stronger noise resistance and can effectively reduce the risk of nonlinear distortion in the transmitting LED. Both theoretical simulations and experiments have proven its superiority.关键词：visible light communication;multiple-input multiple-output;64QAM;superimposed constellation;wireless communication9|127|0更新时间：2024-10-14
- “在图像分类领域，研究者引入自注意力机制至卷积神经网络，显著提升了识别准确率。”摘要：To solve the problem of limited recognition accuracy in image classification tasks on large datasets due to the lack of global information in convolutional neural networks， it is proposed to introduce self attention mechanism into convolutional neural networks. Firstly， image features are extracted through convolutional neural networks and the self attention module is improved； Secondly， the CA module based on convolution operation calculates attention to reconstruct feature maps， highlighting important features and suppressing general features， adding global information to the network； Finally， a Dropout layer is introduced after the Avgpool output layer to reduce overfitting and improve the robustness and generalization performance of the model. Experiments on publicly available datasets ImageNet-1K， Oxford 102 Flowers， and CIFAR-100 have shown that the proposed method improves recognition accuracy by 1.8%， 0.72%， and 13.7% compared to ResNet50， respectively； Compared to the ResNet50 model， it has better recognition performance.关键词：image classification;convolutional neural network;self-attention mechanism;convolution operation;Dropout107|283|0更新时间：2024-10-14
- “在图像隐写领域，提出了基于GAN和分离卷积的高解码精度图像隐写方案，提高了秘密图像安全性、解码精度和编解码速度。”摘要：The problem of low decoding accuracy and image visual quality， long encoding and decoding time are in existing image steganography. In view of the above challenges， a high-decoding accuracy image steganography based on GAN and separated convolution is proposed. An Residual-Rep structure-based and Inception-SCS structure-based preprocessing network is used to automatically learn the high-dimensional features of the image and use the feature representation in a data-driven way before embedding the secret information， acquiring feature information for both channels and spaces， and the skipping connection is used to reduce the loss of secret information， and reduce model complexity by shortening encoding and decoding time. In order to improve the dense decoder’s accuracy， the error correction layer， error correction function and Wasserstein distance are introduced. In a typical environment， an average decoding accuracy of 0.89 and an average structural similarity of 0.95 are obtained， which improves the decoding accuracy and reduces image distortion. The encoding time is reduced by half compared to both SteganoGAN and Hidden methods， allowing the encoding task to be completed in a shorter time.关键词：generative adversarial network;Wasserstein distance;skipping connection;separation convolution;error correction lay23|60|0更新时间：2024-10-14
- “在视频异常检测领域，研究者提出了一种新型深度可分离卷积异常驱动网络，有效降低了计算复杂度，提高了检测性能。”摘要：Video anomaly detection has become a hot issue in current research with profound practical application value. Aiming at the problems of high computational complexity of 3D convolution in video anomaly detection， difficulty in training， and easy overfitting by utilizing only normal data when using reconstruction methods for detection， a novel deeply deparable convolutional anomaly-driven network is proposed. The network firstly extracts jump frames as pseudo-anomaly samples through manual feature extraction to assist training， secondly designs the deeply deparable convolutional network to reduce the number of computational parameters for 3D convolution， and finally allows the network to learn to differentiate between anomalous and normal data by minimizing the reconstruction error of normal data and maximizing the anomalous data. Experimental results show that the model exhibits competitive performance on all major public datasets， with accuracy rates of 91.3%， 99.2%， 87.4% and 98.6% on UCSDped1， UCSDped2， Avenue and UMN datasets， respectively. In addition， the model has strong sensitivity to anomaly detection， and has strong generalization ability and robustness.关键词：deeply separable convolution;pseudo-anomaly;frame reconstruction;video anomalies detection11|75|0更新时间：2024-10-14
- “在自动驾驶领域，研究人员改进了道路标识检测算法，通过双边滤波、Scharr算子和Otsu阈值分割，提高了边缘检测准确性和鲁棒性。”摘要：With the rapid development of autonomous driving technology， road sign detection is becoming increasingly important. To address the issues of edge loss and poor environmental adaptability in current road sign detection， traditional road sign detection algorithms were improved and validated on the OpenCV platform. Firstly， in the image processing stage， an improved bilateral filter is used instead of Gaussian filtering to remove noise and preserve edge information； Then， use the Scharr operator to calculate the gradient amplitude of the image in four directions： x， y， 45 °， and 135 °； Finally， to address the issues of poor threshold adaptation and difficulty in selecting thresholds for the Canny algorithm， the maximum inter class variance （Otsu） method is used for threshold segmentation. Experiments on road sign images have shown that the improved Canny algorithm performs better in single edge response compared to other traditional algorithms， with higher edge detection accuracy and robustness， and relatively shorter processing time.关键词：Canny algorithm;improved bilateral filtering;four-direction Scharr operator;road sign13|96|0更新时间：2024-10-14
- “在微创血管介入导丝检测领域，改进的YOLOv5m网络通过引入可变形卷积和坐标注意力机制，显著提升了检测准确率和速度。”摘要：In order to solve the problems of current mainstream target detection algorithms being less used in minimally invasive vascular interventional guidewire detection， low detection accuracy and slow detection speed， an improved YOLOv5m network is proposed for the detection of vascular interventional guidewires. First， deformable convolution is introduced into the backbone network of YOLOv5m to replace some standard convolutions， and a coordinate attention mechanism is added to the CSP module of the backbone network； BiFPN is used in the neck to performs feature fusion improving the model's ability to fuse different feature layers. Experimental results show that the mAP@0.5 of the improved YOLOv5m algorithm reaches 87.8%， which is 5.7% higher than YOLOv5m， indicating that this algorithm has relatively high application value in vascular interventional guidewire detection.关键词：vascular intervention;YOLOv5m;guidewire detection;deformable convolution;coordinate attention mechanism;BiFPN13|67|0更新时间：2024-10-14
- “在体绘制领域，专家提出了一种高效的空间跳跃加速方法，显著提升了渲染速度。”摘要：With the continuous development of collection technology， a large amount of volume data has been generated in various fields， which usually have complex and unevenly distributed features， posing a huge challenge to the real-time performance of volume rendering. The computational complexity of volume rendering is very high， and it is particularly important to efficiently skip blank space and reduce the number of invalid samples. To this end， an efficient spatial jump volume rendering acceleration method is proposed. Based on the spatial distribution characteristics of volume data， a 3D unsigned distance texture map is pre calculated using a computational shader， which displays the Euclidean distance from the current voxel to the nearest non empty voxel. Performing ray casting in pixel shaders can skip multiple consecutive blank spaces by querying before each sampling， effectively reducing a large number of invalid sampling times. Comparative rendering experiments on various datasets show that the proposed method improves rendering speed by 345% compared to traditional ray casting volume rendering methods， 52% compared to occupation map methods， and 33% compared to Chebyshev distance map methods.关键词：volume rendering;empty space skipping;unsigned distance map;ray casting20|94|0更新时间：2024-10-14
- “在图像翻译领域，研究者基于pix2pix网络提出一种全局信息增强语义图像合成方法，有效提升生成图像质量。”摘要：Semantic image synthesis is an important application and research direction in the field of image translation. Its aim is to generate real images that are consistent with image descriptions using input semantic images， such as semantic segmentation maps， maps and sketches. In response to the problems of blurry image features and lack of correlation in texture details due to the lack of global information in semantic image synthesis tasks based on generative adversarial networks （GANs）， this paper proposes a global information-enhanced semantic image synthesis method based on the pix2pix network model， combined with an external attention mechanism. Firstly， an external attention mechanism is introduced in the upsampling stage of the generator with a U-net structure to enhance the spatial correlation between generated image pixels. Secondly， deep residual modules are used in the upsampling layers of the generator to improve the quality of generated images while enhancing the diversity of the generated images. Finally， the discriminator incorporates global information to enhance its discrimination ability. Experimental evaluations on the Cityscape， Landscape， and Edges2shoes datasets demonstrate the effectiveness of the proposed model. Compared to the baseline model， the improved method achieves improvements of 57.37， 26.74， and 1.78 in terms of the FID （Fréchet Inception Distance） metric for the Cityscape， Landscape， and Edges2shoes datasets， respectively. The results show that the proposed model can effectively utilize global information to enhance the correlation of texture details in generated images and improve the quality of generated images.关键词：image-to-image translation;semantic image synthesis;generative adversarial network;deep learning;computer vision21|90|0更新时间：2024-10-14
- Technical support is provided by Beijing Founder electronics co., LTD 鄂ICP备18021494号-4
鄂公网安备42010602004949号 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0