检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Chen, Jiawei Yang, Xiao Dong, Yinpeng Su, Hang Yin, Zhaoxia Shanghai Key Laboratory of Multidimensional Information Processing East China Normal University China Department of Computer Science & Technology Tsinghua University China

Face anti-spoofing (FAS) and adversarial detection (FAD) have been regarded as critical technologies to ensure the safety of face recognition systems. However, due to limited practicality, complex deployment, and the additional computational overhead, it is necessary to implement both detection techniques within a unified framework. This paper aims to achieve this goal by breaking through two primary obstacles: 1) the suboptimal face feature representation and 2) the scarcity of training data. To address the limited performance caused by existing feature representations, motivated by the rich structural and detailed features of face diffusion models, we propose FaceCat, the first approach leveraging the diffusion model to simultaneously enhance the performance of FAS and FAD. Specifically, FaceCat elaborately designs a hierarchical fusion mechanism to capture rich face semantic features of the diffusion model. These features then serve as a robust foundation for a lightweight head, designed to execute FAS and FAD simultaneously. Due to the limitations in feature representation that arise from relying solely on single-modality image data, we further propose a novel text-guided multi-modal alignment strategy that utilizes text prompts to enrich feature representation, thereby enhancing performance. To combat data scarcity, we build a comprehensive dataset with a wide range of 28 attack types, offering greater potential for a unified framework in facial security. Extensive experiments validate the effectiveness of FaceCat generalizes significantly better and obtains excellent robustness against common input transformations. Copyright © 2024, The Authors. All rights reserved.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Analysis and Design of Image Encryption Algorithms Based on Interlaced Chaos 10th

Analysis and Design of Image Encryption Algorithms Based on ...

引用

10th International Conference on computer Engineering and Networks, CENet 2020

作者： Li, Kangman Li, Qiuping College of Computer Science and Technology Hengyang Normal University Hengyang421002 China Hunan Provincial Key Laboratory of Intelligent Information Processing and Application Hengyang421002 China

ISBN: (纸本)9789811584619

For chaotic encryption algorithm, the algorithm of generating chaotic sequence is complex, and the data is float, which directly affects the speed of encryption. In this paper, an interlaced chaotic encryption algorithm is proposed, In this algorithm, the pixel interval is encrypted by chaotic sequence, and the chaotic sequences of other pixels are obtained by XOR the chaotic values of the front and back points, Experiments show that the algorithm can encrypt and decrypt images quickly and has good effect. © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of Handwritten Digit Recognition Based on K-Nearest Neighbor Algorithm 10th

Design and Implementation of Handwritten Digit Recognition B...

引用

10th International Conference on computer Engineering and Networks, CENet 2020

作者： Wang, Ying Liu, Qingyun Sun, Yaqi Zhang, Feng Zhu, Yining College of Computer Science and Technology Hengyang Normal University Hengyang421002 China Hunan Provincial Key Laboratory of Intelligent Information Processing and Application Hengyang421002 China

ISBN: (纸本)9789811584619

In this paper, a handwritten digit recognition system based on K-nearest neighbor algorithm is designed and implemented. By analyzing the application principle and steps of k-nearest neighbor algorithm in machine learning classifier, we design handwritten digit recognition in Python language. In order to improve the accuracy and accuracy of handwritten character recognition, the method of smooth denoising and character segmentation methods are used to preprocess the data, extract the structure and statistical features. The experiments had shown that the accuracy of the algorithm for handwritten character recognition can reach over 97%. © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Modeling Rumor Unidirectional Spreading from Online Social Networks to Offline 23

Modeling Rumor Unidirectional Spreading from Online Social N...

引用

Proceedings of the 2023 7th International Conference on Electronic information technology and computer Engineering

作者： Qiyi Han Yi Chen Key Laboratory of Pattern Recognition and Intelligent Information Processing School of Computer Science Chengdu University China College of Electronic Engineering Chengdu University of Information Technology China

ISBN: (纸本)9798400708305

Online social networks not only facilitate the dissemination of information, but also increase the risk of rumors. This paper focuses on studying the unidirectional spread of rumors from online social networks to offline environments. To describe the dynamic process of rumor spreading, we derive a unidirectional coupled network structure and mean-field equations. We illustrate the performance of rumor spreading under various scenarios using computer simulations. The simulations reveal that rumors in unidirectional coupled networks spread faster and wider than those in single layer networks. Furthermore, with the assistance of unidirectional links, rumors tend to persist for a longer duration and cause more severe damages.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Modality Unified Attack for Omni-Modality Person Re-Identification

引用

IEEE Transactions on information Forensics and Security 2025年 20卷 5577-5587页

作者： Bian, Yuan Liu, Min Yi, Yunqi Wang, Xueping Ma, Yunfeng Wang, Yaonan Hunan University National Engineering Research Center of Robot Visual Perception and Control Technology College of Electrical and Information Engineering Hunan Changsha China Hunan Normal University Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing College of Information Science and Engineering Hunan Changsha China

Deep learning based person re-identification (re-id) models have been widely employed in surveillance systems. Recent studies have demonstrated that black-box single-modality and cross-modality re-id models are vulnerable to adversarial examples (AEs), leaving the robustness of multi-modality re-id models unexplored. Due to the lack of knowledge about the specific type of model deployed in the target black-box surveillance system, we aim to generate modality unified AEs for omni-modality (single-, cross- and multi-modality) re-id models. Specifically, we propose a novel Modality Unified Attack method to train modality-specific adversarial generators to generate AEs that effectively attack different omni-modality models. A multi-modality model is adopted as the surrogate model, wherein the features of each modality are perturbed by metric disruption loss before fusion. To collapse the common features of omnimodality models, Cross Modality Simulated Disruption approach is introduced to mimic the cross-modality feature embeddings by intentionally feeding images to non-corresponding modality-specific subnetworks of the surrogate model. Moreover, Multi Modality Collaborative Disruption strategy is devised to facilitate the attacker to comprehensively corrupt the informative content of person images by leveraging a multi modality feature collaborative metric disruption loss. Extensive experiments show that our MUA method can effectively attack the omni-modality re-id models, achieving 55.9%, 24.4%, 49.0% and 62.7% mean mAP Drop Rate, respectively. © 2005-2012 IEEE.

关键词： Feature extraction Perturbation methods Measurement Generators Collaboration Robustness Identification of persons Data mining Closed box Training

来源：评论

学校读者我要写书评

暂无评论

Dynamic Student Classification with Forgetting Mechanisms on Memory Networks for Knowledge Tracing

Dynamic Student Classification with Forgetting Mechanisms on...

引用

Intelligent Computing and Human-computer Interaction (ICHCI), International Conference on

作者： Guimin Huang Xiangqian Qin Guangxi Key Laboratory of Image and Graphic Intelligent Processing School of Computer Science and Information Security Guilin University of Electronic Technology Guilin China

ISBN: (数字)9798350368284

ISBN: (纸本)9798350368291

Knowledge Tracing (KT), a technique for modeling students' knowledge levels and predicting their future question-answering performance based on their historical answer data, is one of the key research areas to strengthen the ability of personalized education. In recent years, memory networks have received more and more attention and application in the field of KT, however, the current KT model based on memory networks ignores the effect of students' learning sequence on the level of forgetting, and fails to model the forgetting behavior of students in the process of learning by using the characteristics of time intervals in the interaction data. Therefore, in this paper, we propose a dynamic student classification with forgetting mechanisms on memory networks model (DSCFMN), which enhance the existing model by dynamically classifying students similar to their learning abilities at specified time intervals, incorporating forgetting factors and introducing a weight decay strategy. Experiments show that our model performs well on online education datasets, and the proposed model achieves better prediction results than existing knowledge tracking methods.

关键词： Knowledge engineering Human computer interaction Computational modeling Education Predictive models Data models

来源：评论

学校读者我要写书评

暂无评论

RMSLRS: Real-Time Multi-terminal Sign Language Recognition System 13th

RMSLRS: Real-Time Multi-terminal Sign Language Recognition S...

引用

13th International Conference on Innovations in Bio-Inspired Computing and Applications, IBICA 2022, and 12th World Congress on information and Communication Technologies, WICT 2022

作者： Zhao, Yilin Zhang, Biao Ma, Kun Business School of Jinan University University of Jinan Jinan250022 China School of Computer and Information Hefei University of Technology Hefei230009 China Shandong Provincial Key Laboratory of Network Based Intelligent Computing University of Jinan Jinan250022 China

ISBN: (纸本)9783031274985

Sign language is commonly used by deaf or speech-impaired people to convey meaning. Sign language recognition (SLR) aims to help users learn and use sign language by recognizing the signs from given videos. Although sign language has gained social acceptance, few sign language recognition systems have been developed with educational purposes. Current SLR system has a single form of expression and lacks real-time interaction and feedback with an expensive cost. And videos captured in real scenes are complex, which has impact on the performance of models. To this end, this paper has proposed a novel real-time multi-terminal sign language recognition system (RMSLRS). Specifically, a lightweight sign language recognition model based on MediaPipe Holistic is proposed to perform sign language inference by sensing multi-dimensional information such as pose, facial expression, and hand tracking, achieving near real-time performance on mobile and desktop devices. Then, a novel pre-processing module is proposed to reduce the adverse effects of background noise in videos through the YOLO model and OpenCv. Furthermore, a novel technical architecture for front-end and back-end separation and multi-terminal deployment is designed, including WeChat applet, desktop application and website. Finally, this system has been deployed with gratifying success in practice. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Polarization-Insensitive Orbital Angular Momentum Converter Based on Helically Twisted Anti-Resonant Fiber

Polarization-Insensitive Orbital Angular Momentum Converter ...

引用

2024 Asia Communications and Photonics Conference, ACP 2024 and International Conference on information Photonics and Optical Communications, IPOC 2024

作者： Zhang, Jie Tu, Jiajing Wang, Yingying Gao, Shoufei Huang, Dongmei Gao, Shecheng Liu, Weiping Li, Zhaohui Department of Electronic Engineering Jinan University Guangzhou China College of Physics & Optoelectronic Engineering Jinan University Guangzhou China The Photonics Research Centre Department of Electrical Engineering The Hong Kong Polytechnic University Hung Hom Hong Kong School of Electronics and Information Technology Guangdong Provincial Key Laboratory of Optoelectronic Information Processing Chips and Systems Sun Yat-sen University Guangzhou China

ISBN: (纸本)9798350379266

Based on the topological charge matching principle and mode coupling theory, the higher-order orbital angular momentum (OAM) generating mechanism in a helically-twisted antiresonant fiber (HARF) is illustrated. We successfully fabricate a left-handed HARF by arc discharge method and excite the OAM−1 mode. The experimental result shows that the excitation of the OAM−1 mode is polarization-independent of input light and the polarization of the generated OAM−1 mode can be controlled by changing the polarization of the input light. © 2024 IEEE.

关键词： Angular momentum

来源：评论

学校读者我要写书评

暂无评论

TMFIF:Transformer-based multi-focus image fusion 5

TMFIF:Transformer-based multi-focus image fusion

引用

5th International Conference on computer Vision, Image and Deep Learning, CVIDL 2024

作者： Li, Rui Geng, Shengling Zhang, Dan Zhou, Mingquan Qinghai Normal University School of Computer Science Xining China Nanyang Institute of Technology School of Computer and Software Nanyang China People's Government of Qinghai Province & Beijing Normal University Academy of Plateau Science and Sustainability Xining China Qinghai Normal University The State Key Laboratory of Tibetan Intelligent Information Processing and Application China

ISBN: (纸本)9798350373820

Multi-focus image fusion is a hot topic in the field of image processing, and it is a fundamental problem in the fields of image editing, image synthesis, and target retrieval. In previous fusion methods, although feature-rich datasets, models, and algorithms have been provided, there are still many problems with the effective fusion of distant and near views in complex backgrounds. To solve the challenging multi-focus image fusion problem more accurately, we introduce the Transformer Network based on Encoder Decoder (TMFIF), which can extract more generalized features based on the features of multi-focused images to achieve image fusion in complex backgrounds for better visual effects. In this work, we achieve fusion by inputting two images, near and far view. We compare the performance with other multifocal image fusion algorithms by conducting experiments on publicly available datasets and illustrated using existing evaluation methods and evaluation metrics;the results of the experiments show that our method visualizes the fusion effect better through the encoder and the decoder and the evaluation metrics are also relatively good. © 2024 IEEE.

关键词： Image fusion

来源：评论

学校读者我要写书评

暂无评论

Research on Differential Power Analysis of Lightweight Block Cipher LED 10th

Research on Differential Power Analysis of Lightweight Block...

引用

10th International Conference on computer Engineering and Networks, CENet 2020

作者： Zou, Yi Li, Lang Zhao, Hui-huang Jiao, Ge College of Computer Science and Technology Hengyang Normal University Hengyang421002 China Human Provincial Key Laboratory of Intelligent Information Processing and Application Hengyang421002 China

ISBN: (纸本)9789811584619

LED algorithm is a new lightweight encryption algorithm proposed in CHES 2011, which is used for IOT to protect the communication security of RFID tags and smart cards. It has been found that it is possible to retrieve secret key of algorithm by appropriate analysis of the devicess power consumption, and the differential power analysis is the most powerful attack. This paper proposed a method of differential power analysis on LED algorithm. Combined with hardware circuit, it designed and implemented differential power analysis of the LED encryption system. The experimental results show that method which is proposed in this paper was cracked the 64 bits key of LED. Thus, without protection, the cipher LED would be difficult to resist differential power analysis. and according this, We provide a general analysis method of differential power analysis on other lightweight cryptographic algorithms. © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Light emitting diodes

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：