检索结果-内蒙古大学图书馆

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Xue, Xianglian Jin, Haiyan Zhou, Changjun Xi'an University of Technology School of Computer Science and Engineering Xi'an China Shaanxi University of Chinese Medicine Section of Computer Teaching and Research Xianyang China Shaanxi Key Laboratory for Network Computing and Security Technology Xi'an China Zhejiang Normal University School of Computer Science and Technology Jinhua China

ISBN: (纸本)9798350386226

Medical images often occupy large storage space and contain patient privacy or sensitive information, which makes them difficult and unsafe to be transmitted through the network. This paper proposed an adaptive compression-encryption scheme based on improved compressive sensing (CS) and deoxyribonucleic acid (DNA) coding-compression, which helps to solve the above problem. In the proposed scheme, first the original medical image was compressed to a floating point CS matrix using the discrete wavelet transform and the partial Hadamard matrix. The CS matrix was then normalized and rounded for matrix quantification. The result of quantification was fed to DNA fixed encoding and DNA run length coding for encryption and secondary compression. Finally the compressed-encrypted image was obtained after DNA dynamic encoding-decoding and regroups operation. The proposed scheme was tested against 9 images and proved to be effective in reducing the quantization errors and enhancing the compression performance. For instance, when the benchmark compression ratio (CR) is 0.5, the CR can be reduced by 5%(CR = 0.4453) ∼ 23%(CR = 0.2636), with the corresponding peak signal-to-noise ratio values consistently surpassing the benchmark. Furthermore, the execution of DNA compression and DNA dynamic encoding-decoding provided double guarantee for the algorithm's security. © 2024 IEEE.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

DCGSD: Low-Light Image Enhancement with Dual-Conditional Guidance Sparse Diffusion Model

引用

IEEE Transactions on Circuits and Systems for Video Technology 2025年

作者： Jin, Haiyan Wang, Jing Zuo, Fengyuan Su, Haonan Xiao, Zhaolin Wang, Bin Zhang, Yuanlin Xi'an University of Technology Shaanxi Key Laboratory for Network Computing and Security Technology Department of Computer Science and Engineering NO. 5 South Jinhua Road Shaanxi Xi'an710048 China

When restoring low-light images, most methods largely overlook the ambiguity due to dark noise and lack discrimination for region and shape representations, resulting in invalid feature enhancement. In this work, we propose a physically explainable and prior guidance model for low-light image enhancement, termed Dual-Conditional Guidance Sparse Diffusion (DCGSD). Specifically, we introduce an elaborately designed Luminance Structure Guidance Head, which can be easily plugged into the existing diffusion model to emphasize the value of the luminance and structural representation. Furthermore, for reliable noise analysis, we provide a novel Sparse Attention Enhancement Module that is adaptively empowered to exploit the most useful region-to-region dependencies. This dynamic selection makes the diffusion process from dense to sparse, thus improving the efficiency of the reasoning noise distributions. To avoid noise amplification, we further present a Skip Calibration Module, which can be used to refine the local neighborhood that contains noisy and structural information. Extensive experiments have been performed to verify the superiority of the proposed method. DCGSD shows that leveraging dual-conditional guidance can support the diffusion model to produce sharper and more realistic results. © 1991-2012 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

FIT-Print: Towards False-claim-resistant Model Ownership Verification via Targeted Fingerprint

arXiv

引用

arXiv 2025年

作者： Shao, Shuo Zhu, Haozhe Yao, Hongwei Li, Yiming Zhang, Tianwei Qin, Zhan Ren, Kui State Key Laboratory of Blockchain and Data Security Zhejiang University China Institute of Blockchain and Data Security China Department of Computer Science City University of Hong Kong Hong Kong College of Computing and Data Science Nanyang Technological University Singapore

Model fingerprinting is a widely adopted approach to safeguard the intellectual property rights of open-source models by preventing their unauthorized reuse. It is promising and convenient since it does not necessitate modifying the protected model. In this paper, we revisit existing fingerprinting methods and reveal that they are vulnerable to false claim attacks where adversaries falsely assert ownership of any third-party model. We demonstrate that this vulnerability mostly stems from their untargeted nature, where they generally compare the outputs of given samples on different models instead of the similarities to specific references. Motivated by these findings, we propose a targeted fingerprinting paradigm (i.e., FIT-Print) to counteract false claim attacks. Specifically, FIT-Print transforms the fingerprint into a targeted signature via optimization. Building on the principles of FIT-Print, we develop bit-wise and list-wise black-box model fingerprinting methods, i.e., FIT-ModelDiff and FIT-LIME, which exploit the distance between model outputs and the feature attribution of specific samples as the fingerprint, respectively. Extensive experiments on benchmark models and datasets verify the effectiveness, conferrability, and resistance to false claim attacks of our FIT-Print. © 2025, CC BY.

关键词： Open systems

来源：评论

学校读者我要写书评

暂无评论

Multimodal Multi-objective Particle Swarm Optimization Algorithm Based on Multi-directional Guidance 4

Multimodal Multi-objective Particle Swarm Optimization Algor...

引用

4th International Conference on High Performance Big Data and Intelligent Systems, HDIS 2022

作者： Li, Wei Gao, Yetong Wang, Lei School of Computer Science and Engineering Xi'an University of Technology Xi'an China Shaanxi Key Laboratory for Network Computing and Security Technology Xi'an China

ISBN: (数字)9781665491440

ISBN: (纸本)9781665491440

In multimodal multi-objective optimization, the key issue is to find as many Pareto optimal solutions as possible and select promising solutions in the environmental selection. This paper proposes a multimodal multi-objective particle swarm optimization algorithm based on multi-directional guidance (MM-PSO-MG) to solve these problems. In the proposed algorithm, multi-directional guidance strategy is introduced to avoid premature convergence and find more Pareto optimal solutions. Moreover, the rank-based special crowding distance strategy is used to select promising solutions. 11 multimodal multi-objective test problems are used to verify the performance of the proposed algorithm. The results show that the proposed algorithm is competitive. © 2022 IEEE.

关键词： Multiobjective optimization

来源：评论

学校读者我要写书评

暂无评论

Block-Streaming Service Loading Optimization of Android Dalvik Executable File for Cloud-End Collaboration

Block-Streaming Service Loading Optimization of Android Dalv...

引用

2022 International Conference on Networking and Network Applications, NaNA 2022

作者： Wang, Yichuan Feng, Yanhua Du, Yanning Hei, Xinhong Tian, Yuying Cui, Xiangjie School of Computer Science and Engineering Xi’an University of Technology Xi’an China Shaanxi Key Laboratory for Network Computing and Security Technology Xi’an China

ISBN: (数字)9781665461313

ISBN: (纸本)9781665461313

In recent years, dynamic loading technologies of Android applications have emerged in an endless stream. At present, the existing Android application loading method is to load the entire Android Dalvik Executable (DEX) file for cloud-end collaboration, and there is a problem of finer-grained loading optimization in the loading process. Based on the analysis of Android DEX file in this paper, a formal model of each class is established by using the analysis result of this file, analyzes the user's behavior, and tries to ensure that the class to be used in high probability is loaded into memory first, and a block-streaming service loading optimization model of Android DEX file is put forward. The validity of the model is verified by quantitative analysis of Android DEX file. Trying to load the whole Dalvik Executable file and the classes needed by the application separately, recording and comparing the storage space and loading time overhead. In order to avoid the uncertain factors of the network environment, the above files have been tested and recorded several times. The result shows that the time and storage space for loading the whole DEX file are far greater than the sum of the classes required by the application. With this model, the fine-grained formal analysis of Android DEX file can be performed, which is of great significance for exploring the loading method of application block flow service. © 2022 IEEE.

关键词： Android (operating system)

来源：评论

学校读者我要写书评

暂无评论

Formal Modeling and Defense Methods for 5G Network Endpoint Access Denial of Service Attacks

Formal Modeling and Defense Methods for 5G Network Endpoint ...

引用

2023 International Conference on Networking and Network Applications, NaNA 2023

作者： Zhang, Tong Wei, Wei Wang, Yichuan Deng, Xi Zhu, Lei Ji, Wenjiang School of Computer Science and Engineering Xi'an University of Technology Xi'an China Shaanxi Key Laboratory for Network Computing and Security Technolog Xi'an China School of Computer Science Xi'an Shiyou University Xi'an China

ISBN: (纸本)9798350327380

5G terminal access denial-of-service attack is one of the more common types of 5G signaling network attacks. It manifests itself as a large number of abnormal access requests occupying core network resources. Normal requests cannot obtain resources, resulting in unstable network services, and users will be unable to connect or drop connections, eventually leading to frequent terminal takeoffs and landings. This paper proposes a formal modeling scheme for a 5G terminal access denial-of-service attack. The attack principle is analyzed, a defense scheme is designed and modeled, and analyzed, and then quantitative experiments are conducted, and the experimental results show the effectiveness of the defense scheme. © 2023 IEEE.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Membership Data Privacy Protection and Poisoning Detection Scheme for Federated Learning 1st

Membership Data Privacy Protection and Poisoning Detection ...

引用

1st Conference on blockchain and Web3.0 Technology Innovation and Application, BWTAC 2024

作者： Li, Yafeng Sun, Zhijun Ma, Lichuan China CETC Key Laboratory of Technology on Data Link Xi’an710068 China State Key Laboratory of Integrated Services Networks Xidian University Xi’an710071 China Shaanxi Key Laboratory of Blockchain and Secure Computing Xi’an710071 China

ISBN: (纸本)9789819794119

In the era of big data, data privacy and security are highly valued, leading to the phenomenon of "data islands", which hinders the effective utilization of data. Federated learning, as a privacy-preserving distributed machine learning method, allows model training without sharing raw data, addressing the privacy concerns in cross-organizational data cooperation. Although federated learning protects data privacy, participants’ gradient parameters can still potentially leak sensitive information. As a result, extensive research has focused on conducting federated learning while protecting the privacy of gradient parameters and reducing the communication and computational overhead for participants. These efforts assume that each participant adheres honestly to the protocol;however, malicious submissions of incorrect gradient parameters can significantly bias the federated learning model. This paper proposes a federated learning scheme for privacy-protected gradient aggregation and poison detection that defends against malicious poisoning attacks while safeguarding data privacy, and optimizes computational and communication overheads. By employing inner product triples, the computation process is further optimized, enhancing system performance. The security of the scheme is demonstrated, and its efficiency is validated through practical tests, showing that the optimized solution significantly outperforms the basic scheme in terms of performance. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Precise and Efficient:A Hybrid Analysis-Based Stack Overflow Detection Model under MIPS 9

Precise and Efficient:A Hybrid Analysis-Based Stack Overflow...

引用

9th IEEE Smart World Congress, SWC 2023

作者： Wang, Yichuan Huang, Chenhao Zhang, Tong Zhao, Li Gu, Zhendong Li, Gangyang Xi'an University of Technology Xi'an China Shaanxi Key Laboratory for Network Computing and Security Technology China Hanjiang.to.Weihan River Valley Water Diversion Project Construction.co.ltd Xi'an China

ISBN: (纸本)9798350319804

Although the vulnerability detection techniques under the MIPS platform can detect stack overflow vulnerabilities, the stack overflow vulnerability of the stack frame changes is complex and hard to recover. The input generated by fuzzy testing techniques is inefficient and difficult to recover the site after generating crashes. It is a challenge to preserve the context of runtime stack and establish effective constraints under hybrid analysis techniques, which usually makes the solution path more run inefficient and affects the analysis results in accuracy. This paper proposes a method that combines hybrid analysis including taint analysis to detect the vulnerabilities of stack overflow. The method ensures accuracy while reducing the number of execution paths. The three vulnerabilities of stack overflow in the routed firmware selected for testing are successfully detected and the number of detected paths is controlled to about 10. In addition, the method is designed to decouple the vulnerability location and detection modules, providing a basis for subsequent research on a common vulnerability detection platform. © 2023 IEEE.

关键词： binary analysis binary security embedded security MIPS vulnerability discovery

来源：评论

学校读者我要写书评

暂无评论

Cross Elitist Learning Multifactorial Evolutionary Algorithm 1

引用

3rd International Conference on Neural computing for Advanced Applications, NCAA 2022

作者： Li, Wei Luo, Haonan Wang, Lei School of Computer Science and Engineering Xi’an University of Technology Xi’an710048 China Shaanxi Key Laboratory for Network Computing and Security Technology Xi’an710048 China

ISBN: (数字)9789811961427

ISBN: (纸本)9789811961410

Multifactorial Evolutionary Algorithm (MFEA) is a popular optimization algorithm in recent years. It has implicit parallelism and can solve different problems at the same time in the same search space. However, premature convergence is a significant shortcoming of MFEA. The main reason for this phenomenon is the lack of population diversity and the inability to migrate effectively between tasks. To address this shortcoming, this paper proposes a cross elitist learning multifactorial evolutionary algorithm (CEL-MFEA). Firstly, the cross elitist strategy guides the individuals to learn from the elite individual, which can improve the convergence of the proposed CEL-MFEA. Secondly, the Nelder-Mead algorithm is used to provide an effective knowledge transfer between different tasks. Finally, the opposition learning mechanism is employed to ensure population diversity. The comprehensive ability of the proposed CEL-MFEA is evaluated by 9 classical multifactorial optimization problem test sets. Experimental results show that the CEL-MFEA is more competitive than MFEA algorithm and several popular multifactorial optimization algorithms. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

Adaptive Region Sampling Network For Polarimetric SAR Image Classification

Adaptive Region Sampling Network For Polarimetric SAR Image ...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Junfei Shi Shanshan Ji Haiyan Jin Haonan Su Zhiyong Lv Department of Computer Science and Technology Shaanxi Key Laboratory for Network Computing and Security Technology Xi’an University of Technology Xi’an China

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Deep learning models have demonstrated excellent performance for polarimetric SAR image classification. However, existing approaches generally use a fixed square window to sample image blocks as the network input, which may not effectively extract various terrain objects. To alleviate this issue, we proposed an adaptive region sampling network to learn different terrain types by introducing a novel sampling scheme with varying direction and scale. Initially, a complex PolSAR image is segmented into homogeneous, heterogeneous and boundary regions. Subsequently, small-scale and large-scale sampling windows are designed for homogeneous and heterogeneous regions, to capture local and global features for two types of regions respectively. Additionally, an adaptive directional sampling window is designed for boundary regions to ensure context consistency in the image block and prevent edge confusion. Experiments conducted on real PolSAR data sets demonstrate that our method achieves superior classification results, providing both regional consistency and boundary preservation.

关键词： Representation learning Deep learning Image segmentation Adaptive systems Accuracy Image edge detection Semantics

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：