检索结果-内蒙古大学图书馆

WFSS: weighted fusion of spectral transformer and spatial self-attention for robust hyperspectral image classification against adversarial attacks

引用

Visual Intelligence 2024年第1期2卷 1-16页

作者： Tang, Lichun Yin, Zhaoxia Su, Hang Lyu, Wanli Luo, Bin Anhui Provincial Key Laboratory of Multimodal Cognitive Computation Anhui University Anhui China School of Communication & Electronic Engineering East China Normal University Shanghai China Department of Computer Science and Technology Tsinghua University Beijing China

The emergence of adversarial examples poses a significant challenge to hyperspectral image (HSI) classification, as they can attack deep neural network-based models. Recent adversarial defense research tends to establish global connections of spatial pixels to resist adversarial attacks. However, it cannot yield satisfactory results when only spatial pixel information is used. Starting from the premise that the spectral band is equally important for HSI classification, this paper explores the impact of spectral information on model robustness. We aim to discover potential relationships between different spectral bands and establish global connections to resist adversarial attacks. We design a spectral transformer based on the transformer structure to model long-distance dependency relationships among spectral bands. Additionally, we use a self-attention mechanism in the spatial domain to develop global relationships among spatial pixels. Based on the above framework, we further explore the influence of both spectral and spatial domains on the robustness of the model against adversarial attacks. Specifically, a weighted fusion of spectral transformer and spatial self-attention (WFSS) is designed to achieve the multi-scale fusion of spectral and spatial connections, which further improves the model’s robustness. Comprehensive experiments on three benchmarks show that the WFSS framework has superior defensive capabilities compared to state-of-the-art HSI classification methods. © The Author(s) 2024.

关键词： Adversarial attack Adversarial example Hyperspectral image (HSI) classification Self-attention Transformer

来源：评论

学校读者我要写书评

暂无评论

PR3D: Precise and realistic 3D face reconstruction from a single image

PR3D: Precise and realistic 3D face reconstruction from a si...

引用

作者： Huang, Zhangjin Wu, Xing University of Science and Technology of China Hefei China Anhui Province Key Laboratory of Software in Computing and Communication Hefei China Deqing Alpha Innovation Institute Huzhou China

Reconstructing the three-dimensional (3D) shape and texture of the face from a single image is a significant and challenging task in computer vision and graphics. In recent years, learning-based reconstruction methods have exhibited outstanding performance, but their effectiveness is severely constrained by the scarcity of available training data with 3D annotations. To address this issue, we present the PR3D (Precise and Realistic 3D face reconstruction) method, which consists of high-precision shape reconstruction based on semi-supervised learning and high-fidelity texture reconstruction based on StyleGAN2. In shape reconstruction, we use in-the-wild face images and 3D annotated datasets to train the auxiliary encoder and the identity encoder, encoding the input image into parameters of FLAME (a parametric 3D face model). Simultaneously, a novel semi-supervised hybrid landmark loss is designed to more effectively learn from in-the-wild face images and 3D annotated datasets. Furthermore, to meet the real-time requirements in practical applications, a lightweight shape reconstruction model called fast-PR3D is distilled through teacher–student learning. In texture reconstruction, we propose a texture extraction method based on face reenactment in StyleGAN2 style space, extracting texture from the source and reenacted face images to constitute a facial texture map. Extensive experiments have demonstrated the state-of-the-art performance of our method. © 2024 John Wiley & Sons Ltd.

关键词： Textures

来源：评论

学校读者我要写书评

暂无评论

Adversarial Example Defense via Perturbation Grading Strategy 9th

Adversarial Example Defense via Perturbation Grading Strate...

引用

9th International Forum on Digital Multimedia communication, IFTC 2022

作者： Zhu, Shaowei Lyu, Wanli Li, Bin Yin, Zhaoxia Luo, Bin Anhui Provincial Key Laboratory of Multimodal Cognitive Computation Anhui University Hefei China Guangdong Key Laboratory of Intelligent Information Processing and Shenzhen Key Laboratory of Media Security Shenzhen University Shenzhen China School of Communication and Electronic Engineering East China Normal University Shanghai China

ISBN: (纸本)9789819908554

Deep Neural Networks have been widely used in many fields. However, studies have shown that DNNs are easily attacked by adversarial examples, which have tiny perturbations and greatly mislead the correct judgment of DNNs. Furthermore, even if malicious attackers cannot obtain all the underlying model parameters, they can use adversarial examples to attack various DNN-based task systems. Researchers have proposed various defense methods to protect DNNs, such as reducing the aggressiveness of adversarial examples by preprocessing or improving the robustness of the model by adding modules. However, some defense methods are only effective for small-scale examples or small perturbations but have limited defense effects for adversarial examples with large perturbations. This paper assigns different defense strategies to adversarial perturbations of different strengths by grading the perturbations on the input examples. Experimental results show that the proposed method effectively improves defense performance. In addition, the proposed method does not modify any task model, which can be used as a preprocessing module, which significantly reduces the deployment cost in practical applications. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Grading

来源：评论

学校读者我要写书评

暂无评论

How Can We Cope with the Impact of Microservice Architecture Smells? 2022

How Can We Cope with the Impact of Microservice Architecture...

引用

11th International Conference on software and Computer Applications, ICSCA 2022

作者： Ding, Xiang Zhang, Cheng Anhui Provincial Key Laboratory of Multimodal Cognitive Computation Anhui University China School of Computer Science and Technology Anhui University China State Key Lab for Novel Software Technology Nanjing University China

ISBN: (纸本)9781450385770

Context: software Architecture Smells (AS) are potential software structure problems and always impact software quality negatively. And with the development of Microservice architecture, the Microservice Architecture Smells (MAS) have been attracting more and more attention. The software scholars and developers have discovered the influence of MAS and performed some researches on them. However, the definition and category of MAS are still ambiguous. Objects: This paper aims to clarify the specific MAS categories and their definitions, and then further explores the issues caused by MAS in the migration process from a monolithic system to Microservice. Method: We performed a comprehensive systematic literature review about MAS. Specifically, we explored 13 white and 10 grey literature in detail to get MAS information by using the quantitative research method. To explore the issues that influence the migration process from a monolithic system to Microservice, we used the meta-ethnography qualitative research method to extract relevant information and get six three-order translations. Results: This study defined 22 Microservice Architecture Smells and classified them into five categories, namely Design, Deployment, Monitor & Log, communication and Team & Tool, based on their characteristics. Simultaneously, the issues that influence the migration process are proposed, including service cutting, databases, communication, team and techniques. Finally, we matched the MAS to the issues they caused in the migration process and recommended solutions to these issues. © 2022 ACM.

关键词： meta-ethnography Microservice Architecture Smells smell impact Systematic Literature Review

来源：评论

学校读者我要写书评

暂无评论

Implementation of Domain-oriented Microservices Decomposition based on Node-attributed Network 2022

Implementation of Domain-oriented Microservices Decompositio...

引用

11th International Conference on software and Computer Applications, ICSCA 2022

作者： Cao, Lingli Zhang, Cheng Anhui Provincial Key Laboratory of Multimodal Cognitive Computation Anhui University China School of Computer Science and Technology Anhui University China State Key Lab for Novel Software Technology Nanjing University China

ISBN: (纸本)9781450385770

The features of microservices, such as scalability and maintainability, have attracted the industry to migrate monolithic projects to microservices. However, how to decompose microservices during migration is a tricky problem. At present, microservices decomposition mainly relies on architects or domain experts, which is more subjective and time-consuming. Followed by semi-automated or automated microservice decomposition, such methods produce coarse-grained results affected by different system characteristics, which cannot make desired decomposition according to the migration requirements of domains. Therefore, this paper proposes a domain-oriented fine-grained microservices decomposition resolution scheme. It uses dynamic and static analysis to obtain the invocation relationships and invocation times between entity methods and the response time of entity methods to represent three main system characteristics concerned during the migration: function, inter-service communications, and performance. And express the above information of monolith by the node-attributed network. Then it uses the community detection algorithm and the proposed similar hierarchical clustering algorithm to complete objective and effective decomposition. Finally, the rationality and feasibility of the proposed approach are verified using the JPetStore case. © 2022 ACM.

关键词： Dynamic analysis Static analysis Microservices Decomposition Node-attributed Network Domain

来源：评论

学校读者我要写书评

暂无评论

Multi-objective software test case selection based on density analysis 22

Multi-objective software test case selection based on densit...

引用

5th International Conference on Computer Science and software Engineering, CSSE 2022

作者： Jia, Huihui Zhang, Cheng Wu, Sijie Anhui Provincial Key Laboratory of Multimodal Cognitive Computation Anhui University China School of Computer Science and Technology Anhui University China State Key Lab for Novel Software Technology Nanjing University Hefei China

ISBN: (纸本)9781450397780

software test case selection is committed to select the fewest test cases from test suites to perform a complete test at the least cost. Machine learning and multi-objective optimization techniques have developed rapidly in recent years, and they have been successfully applied to test case selection. In this paper, we present a method called DB-NSGA2, which uses the density clustering algorithm in machine learning combined with the non-dominated ranking algorithm (NSGA2) for test case selection, which can better select the test cases required for testing. In particular, we apply some of the clustering results generated by the clustering algorithm to the crossover and mutation operations of the NSGA2 to improve diversity progeny populations and ensure the transmission of good individuals. Extensive experiments show that the test cases selected by our method can produce a better set of Pareto solutions and can detect more faults at a lower cost than other methods. © 2022 ACM.

关键词： Clustering algorithms

来源：评论

学校读者我要写书评

暂无评论

Identification of Microservices through Processed Dynamic Traces and Static Calls 3

Identification of Microservices through Processed Dynamic Tr...

引用

3rd International Conference on Computer Science and Management Technology, ICCSMT 2022

作者： Wu, Sijie Zhang, Cheng Anhui Provincial Key Laboratory of Multimodal Cognitive Computation Anhui University China Anhui University School of Computer Science and Technology China Nanjing University State Key Lab for Novel Software Technology Hefei China

ISBN: (纸本)9798350320404

As the complexity of software systems continues to increase, traditional monolithic software architectures will become difficult to scale and maintain. Microservices as an emerging software architecture it advocates decentralized management, service autonomy, and distinct module boundaries. These features of microservices make software systems easy to maintain, deploy, and modify. However, microservice identification is still not well addressed due to the lack of an effective approach. The existing work on microservice identification mainly starts from the system functionality perspective and does not take into account the actual business rules, which leads to microservice candidates not conforming to the concept of microservices. To address this challenge, we propose an automated approach to microservice identification that combines dynamic and static analysis and takes into account the actual business. Our proposed approach starts with a dynamic and static analysis of the monolithic system. Once the behavioral data of the system is collected, the behavioral data will be further processed according to our proposed guidelines. Then, the proposed approach calculates the degree of the correlation between classes in a monolithic system. Finally, we propose an improved clustering algorithm based on K-means to classify the classes in a monolithic system into different microservices. The proposed approach is validated by comparison with other approaches on open source projects. © 2022 IEEE.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

An Innovative Approach To software Modularization Based On The Artificial Fish Swarm Algorithm 22

An Innovative Approach To Software Modularization Based On T...

引用

Proceedings of the 2022 6th International Conference on Electronic Information Technology and Computer Engineering

作者： Jianqiang Pan Cheng Zhang Huihui Jia Anhui Provincial Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology State Key Lab for Novel Software Technology Anhui University China

ISBN: (纸本)9781450397148

It gets more and more expensive to maintain the complete software system as time goes on since the software architecture grows more complicated and the software code is more difficult to understand. This issue may be solved by taking the essential parts out of the source code and organizing them into the appropriate subsystems. Hierarchy-based partitioning is more complex and less successful in solving issues with huge software modules. In order to do this, this work suggests an innovative artificial fish swarm method as a meta-heuristic to solve the software module clustering problems (SMCPs) and presents a mutation operator and a local search strategy to handle the premature situation. The performance of the suggested method is assessed on a range of real-world software systems by comparison to the most recent meta-heuristic methods. The outcomes of the experiments demonstrate that the suggested methodology works better than those of other methods.

关键词： software refactoring Artificial fish swarm algorithm software module clustering

来源：评论

学校读者我要写书评

暂无评论

Linguistic Steganalysis Based on Meta Features and Immune Mechanism

引用

Chinese Journal of Electronics 2023年第4期19卷 661-666页

作者： Hao Yang Xianbin Cao Anhui Province Key Laboratory of Software in Computing and Communication School of Computer Science and Technology University of Science and Technology of China Hefei China

Linguistic steganalysis depends on efficient detection features due to the diversity of syntax and the polysemia of semantics in natural language processing. This paper presents a novel linguistics steganalysis approach based on meta features and immune clone mechanism. Firstly, meta features are used to represent texts. Then immune clone mechanism is exploited to select appropriate features so as to constitute effective detectors. Our approach employed meta features as detection features, which is an opposite view from the previous literatures. Moreover, the immune training process consists of two phases which can identify respectively two kinds of stego texts. The constituted detectors have the capable of blind steganalysis to a certain extent. Experiments show that the proposed approach gets better performance than typical existing methods, especially in detecting short texts. When sizes of texts are confined to 3kB, detection accuracies have exceeded 95%.

关键词： Training Integrated circuits Steganography Semantics Cloning Detectors Linguistics

来源：评论

学校读者我要写书评

暂无评论

Tumor type classification and candidate cancer-specific biomarkers discovery via semi-supervised learning

引用

Biophysics Reports 2023年第2期9卷 57-66页

作者： Peng Chen Zhenlei Li Zhaolin Hong Haoran Zheng Rong Zeng School of Computer Science and Technology University of Science and Technology of China Anhui Key Laboratory of Software Engineering in Computing and Communication University of Science and Technology of China Department of Systems Biology University of Science and Technology of China CAS Key Laboratory of Systems Biology Shanghai Institute of Biochemistry and Cell Biology Center for Excellence in Molecular Cell Science Chinese Academy of Sciences School of Life Science and Technology ShanghaiTech University

Identifying cancer-related differentially expressed genes provides significant information for diagnosing tumors, predicting prognoses, and effective treatments. Recently, deep learning methods have been used to perform gene differential expression analysis using microarray-based high-throughput gene profiling and have achieved good results. In this study, we proposed a new robust multiple-datasetsbased semi-supervised learning model, MSSL, to perform tumor type classification and candidate cancer-specific biomarkers discovery across multiple tumor types and multiple datasets, which addressed the following long-lasting obstacles:(1) the data volume of the existing single dataset is not enough to fully exert the advantages of deep learning;(2) a large number of datasets from different research institutions cannot be effectively used due to inconsistent internal variances and low quality;(3) relatively uncommon cancers have limited effects on deep learning methods. In our article, we applied MSSL to The Cancer Genome Atlas(TCGA) and the Gene Expression Comprehensive Database(GEO) pan-cancer normalized-level3 RNA-seq data and got 97.6% final classification accuracy, which had a significant performance leap compared with previous approaches. Finally, we got the ranking of the importance of the corresponding genes for each cancer type based on classification results and validated that the top genes selected in this way were biologically meaningful for corresponding tumors and some of them had been used as biomarkers, which showed the efficacy of our method.

关键词： Deep learning Tumor type classification Cancer-specific biomarkers MSSL

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：