检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Ning-Qian Wu Ya-Jun Hu Liping Chen Zhen-Hua Ling National Engineering Research Center of Speech and Language Information Processing University of Science and Technology of China Hefei P.R.China iFLYTEK Research iFLYTEK Co. Ltd. China MoE Key Laboratory of Brain-Inspired Intelligent Perception and Cognition University of Science and Technology of China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Spontaneous behaviors in speech pose significant challenges for speech synthesis. Existing research has not adequately addressed these behaviors, with most studies relying on specially recorded datasets. In contrast, real-world data more accurately reflects the natural, spontaneous speaking styles in everyday life and encompasses a wider range of spontaneous behaviors. However, such data is often of lower quality, and the distribution of spontaneous behaviors is highly imbalanced. In this study, we explore spontaneous speech synthesis using real-world data within the VITS2 framework. To overcome these challenges, we introduce two techniques: anchored monotonic alignment and spontaneous hidden representation substitution. Experimental results demonstrate that these methods enhance model alignment and improve the naturalness of the generated speech. Our proposed approach successfully addresses the challenge of synthesizing rare spontaneous behaviors and offers users flexible control over the synthesized speech.

关键词： Training Accuracy Large language models Speech enhancement Signal processing Acoustics Speech synthesis

来源：评论

学校读者我要写书评

暂无评论

Air Pollution Monitoring By Integrating Local and Global information in Self-Adaptive Multiscale Transform Domain

引用

IEEE Transactions on Multimedia 2025年

作者： Gu, Ke Liu, Yuchen Liu, Hongyan Liu, Bo Qiao, Junfei Lin, Weisi Zhang, Wenjun Beijing University of Technology School of Information Science and Technology Engineering Research Center of Intelligent Perception and Autonomous Control of Ministry of Education Beijing Laboratory of Smart Environmental Protection Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing Artificial Intelligence Institute Beijing100124 China Massey University School of Mathematical and Computational Sciences New Zealand Nanyang Technological University College of Computing and Data Science Singapore Shanghai Jiao Tong University Institute of Image Communication and Information Processing Shanghai China

As the deadliest form of pollution, air pollution had a prolonged severe damage to the human health and life safety of nearly 99% of the world's population. Facing to the problem that billions of tons of pollutants discharged into the air per year by industrial activities, this paper proposed a novel image-based air pollution monitor (IAPM) by incorporating local and global information in the self-adaptive multiscale transform domain, so as to achieve the timely and effective leakage detection of typical air pollutants from a single image. To be specific, this paper first developed a screen-shaped module according to two significant findings in visual neuroscience, which include the high sensitivity of human eyes to horizontal and vertical stimuli and the center-surround inhibition, by designing and fusing the square module, horizontal strip module and vertical strip module parallelly for simulating the behaviour of human eyes to extract local features. Second, the learnable weights and proportional mapping were applied to incorporate the screen-shaped module and lightweight vision transformer as backbone, towards more richly exploiting and fusing local and global information just as the way a brain perceives external stimuli. Third, a new self-adaptive multiscale transform domain method was devised based on two motivations from the visual characteristics of multiscale perception and the brain characteristics of self-adaptive domain transform to modify the backbone by using the operations of pooling and pointwise convolution. Extensive experiments implemented on the datasets of carbon particulate matters and ethylene leakage confirmed the superior monitoring performance of the proposed IAPM model beyond the state-of-the-art (SOTA) peers by an accuracy gain of about 4%. Furthermore, the proposed IAPM model only required 0.089 GFLOPs and 0.15 million model parameters, remarkably outperforming SOTA competitors in computational efficiency and storage resources. © 1999-2

关键词： information leakage

来源：评论

学校读者我要写书评

暂无评论

ALCencryption:A Secure and Efficient Algorithm for Medical Image Encryption

引用

Computer Modeling in Engineering & Sciences 2020年第12期125卷 1083-1100页

作者： Jiao Ge College of Computer Science and Technology Hengyang Normal UniversityHengyang421002China Hunan Provincial Key Laboratory of Intelligent Information Processing and Application Hengyang421002China

With the rapid development of medical informatization and the popularization of digital imaging equipment,DICOM images contain the personal privacy of patients,and there are security risks in the process of storage and transmission,so it needs to be *** order to solve the security problem of medical images on mobile devices,a safe and efficient medical image encryption algorithm called ALCencryption is *** algorithm first analyzes the medical image and distinguishes the color image from the gray *** gray images,the improved Arnold map is used to scramble them according to the optimal number of iterations,and then the diffusion is realized by the Logistic and Chebyshev map cross-diffusion *** color image is encrypted by cross-diffusion algorithm of double chaotic *** and efficiency analysis show that the ALCencryption algorithm has the characteristics of small neighboring pixels,large key space,strong key sensitivity,high safety and short encryption *** is suitable for medical image encryption of mobile devices with high real-time requirements.

关键词： Patient privacy DICOM medical image encryption scrambling degree cross-diffusion

来源：评论

学校读者我要写书评

暂无评论

The interior penalty virtual element method for fourth-order singular perturbation problems

arXiv

引用

arXiv 2023年

作者： Feng, Fang Yu, Yue School of Mathematics and Statistics Nanjing University of Science and Technology Nanjing China School of Mathematics and Computational Science Hunan Key Laboratory for Computation and Simulation in Science and Engineering Key Laboratory of Intelligent Computing and Information Processing Ministry of Education Xiangtan University Hunan Xiangtan411105 China

This paper is dedicated to the numerical solution of a fourth-order singular perturbation problem using the interior penalty virtual element method (IPVEM) proposed in [42]. The study introduces modifications to the jumps and averages in the penalty term, as well as presents an automated mesh-dependent selection of the penalty parameter. Drawing inspiration from the modified Morley finite element methods, we leverage the conforming interpolation technique to handle the lower part of the bilinear form. Through our analysis, we establish optimal convergence in the energy norm and provide a rigorous proof of uniform convergence concerning the perturbation parameter in the lowest-order case. © 2023, CC BY-NC-ND.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

GLFAFormer: DeepFake forgery detection with adaptive feature extract and align

引用

Digital Signal processing: A Review Journal 2025年 165卷

作者： Chen, Long Zhao, Huihuang He, Jiaxin Meng, Weiliang College of Computer Science and Technology Hengyang Normal University Hengyang421002 China Hunan Provincial Key Laboratory of Intelligent Information Processing and Application Hengyang421002 China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing100049 China

With the rapid development of generative models, there is an increasing demand for universal fake image detectors. In this paper, we investigate the problem of fake image detection for the synthesis of generative models to detect fake images from multiple generative methods. Recent research methods explore the benefits of pre-trained models and mainly adopt a fixed paradigm of training additional classifiers separately, but we find that the fixed paradigm hinders the full learning of forgery features, leading to insufficient representation learning in the detector. The main reason is that the fixed paradigm pays too much attention to global features and neglects local features, which limits the ability of the model to perceive image details and leads to some information loss or confusion. In this regard, based on the pre-trained visual language space, our method introduces two core designs. First, we designed a Deep Window Triple Attention (DWTA) module. A similar dense sliding window strategy is adopted to capture multi-scale local abnormal patterns, and the sensitivity to generated artifacts is enhanced through the triple attention mechanism. Secondly, we proposed a Cross-Space Feature Alignment (CSFA) module. A two-way interactive channel between global features and local features is established, and the alignment loss function is used to achieve semantic alignment of cross-modal feature spaces. The aligned features are adaptively fused through a gating mechanism to obtain the final adaptive forged features. Experiments demonstrate that our method, when trained solely on ProGAN data, achieves superior cross-generator generalization: it attains an average accuracy of 94.7% on unseen GANs and generalizes to unseen diffusion models with 94% accuracy, surpassing existing methods by 2.1%. The source code is available at https://***/long2580h/GLFAFormer. © 2025 Elsevier Inc.

关键词： Visual languages

来源：评论

学校读者我要写书评

暂无评论

Structure-constrain Source Free Domain Adaptation with Regularized Knowledge Distillation of Source Model

Structure-constrain Source Free Domain Adaptation with Regul...

引用

Data Driven Control and Learning Systems (DDCLS)

作者： Dan Yang Xin Peng Cheng Su Dayu Tan Chaoyang Chen Weimin Zhong Key Laboratory of Smart Manufacturing in Energy Chemical Process Ministry of Education East China University of Science and Technology Shanghai China Key Laboratory of Intelligent Computing and Signal Processing Ministry of Education Anhui University Hefei Anhui China School of Information and Electrical Engineering Hunan University of Science and Technology Xiangtan China

ISBN: (数字)9798350361674

ISBN: (纸本)9798350361681

Transfer learning aims to leverage the knowledge of the source domain to help the learning of unlabeled target domain models. However, all conventional transfer learning methods assume that samples from source and target domains are always freely accessible. However, this assumption is rarely plausible in real-world due to data privacy. To combat this challenge, we introduce Structure-constrain Source Free Domain Adaptation with regularized Knowledge Distillation of source model (SSFDA-KD). Our method is based on the observations that a deterministic classification result can be obtained when the target data distribution aligns with the source data distribution, and a well-trained model not only yields a deterministic classification for a sample but also has clear category boundaries, which implies samples that are close together belong to the same category. From those, we first generate the source model with labeled source data, which is regarded as a well-trained model. Then, we identify target prototypes with lower self-entropy and assign the pseudo-labels to other samples according to the neighborhood structure. Finally, we refine the target model by minimizing prediction loss, constraining internal structure via triplet loss, and reducing knowledge distillation loss. Experiments of fault diagnosis based on WWTPs are conducted to validate its effectiveness.

关键词： Fault diagnosis Learning systems Adaptation models Data privacy Transfer learning Prototypes Predictive models

来源：评论

学校读者我要写书评

暂无评论

A posteriori error estimation for an interior penalty virtual element method of Kirchhoff plates

arXiv

引用

arXiv 2024年

作者： Feng, Fang Hu, Yuming Yu, Yue Zhao, Jikun School of Mathematics and Statistics Nanjing University of Science and Technology Nanjing China Hunan Key Laboratory for Computation and Simulation in Science and Engineering Key Laboratory of Intelligent Computing and Information Processing of Ministry of Education National Center for Applied Mathematics in Hunan School of Mathematics and Computational Science Xiangtan University Hunan Xiangtan411105 China School of Mathematics and Statistics Zhengzhou University Zhengzhou450001 China

In this paper, we develop a residual-type a posteriori error estimation for an interior penalty virtual element method (IPVEM) for the Kirchhoff plate bending problem. Building on the work in [18], we adopt a modified discrete variational formulation that incorporates the H1-elliptic projector in the jump and average terms. This allows us to simplify the numerical implementation by including the H1-elliptic projector in the computable error estimators. We derive the reliability and efficiency of the a posteriori error bound by constructing an enriching operator and establishing some related error estimates that align with C0-continuous interior penalty finite element methods. As observed in the a priori analysis, the interior penalty virtual elements exhibit similar behaviors to C0continuous elements despite its discontinuity. This observation extends to the a posteriori estimate since we do not need to account for the jumps of the function itself in the discrete scheme and the error estimators. As an outcome of the error estimator, an adaptive VEM is introduced by means of the mesh refinement strategy with the one-hanging-node rule. Numerical results from several benchmark tests confirm the robustness of the proposed error estimators and show the efficiency of the resulting adaptive VEM. Copyright © 2024, The Authors. All rights reserved.

关键词： Discrete element methods

来源：评论

学校读者我要写书评

暂无评论

Dual-Stream Fusion Network With Multi-Head Self-Attention for Multi-Modal Fake News Detection

SSRN

引用

SSRN 2024年

作者： Yang, Yimei Liu, Jinping Yang, Yujun Cen, Lihui School of Mathematics and Statistics Hunan Normal University Changsha410081 China Huaihua University Hunan Huaihua418000 China College of Information Science and Engineering Hunan Normal University Changsha China School of Automation Central South University Changsha410083 China Key Laboratory of Wuling-Mountain Health Big Data Intelligent Processing Application in Hunan Province Universities Hunan Huaihua418000 China Key Laboratory of Intelligent Control Technology for Wuling-Mountain Ecological Agriculture in Hunan Province Hunan Huaihua418000 China

With the rapid development of social media platforms such as Weibo, WeChat, and other new media, as well as the rise of deepfake technologies, the identification and containment of fake information have posed significant challenges, becoming one of the key challenges faced by society and governmental institutions. Therefore, the development of intelligent and efficient methods for detecting fake news has become a top priority in combating misinformation. Addressing challenges in current fake news detection techniques, such as cross-lingual issues, data imbalance, and multimodal information fusion, this paper aims to propose a dual-stream fusion network model (DSF-MHSA) based on deep learning for detecting fake news across web pages, images, and text. The model combines the deep learning models ERNIE-M, AlexNet, ShuffleNet, and three multi-head self-attention mechanisms to extract and classify features from multilingual text and images. By utilizing two separate multi-head self-attention mechanisms to process extracted textual and image information, the model captures long-range dependencies among texts to better understand contextual information and captures global information among images to enhance object recognition and localization capabilities. Finally, a single multi-head self-attention mechanism is employed to integrate textual and image information to capture cross-modal correlations, thereby strengthening the model's understanding and identification capabilities of news content. This dual-stream fusion network architecture ensures the model's ability to detect multimodal news scenarios in both English and Chinese short texts and images. Experimental evaluations conducted on three publicly available datasets - Twitter, Weibo, and IKCEST-2023 dataset, which contains real online news data collected from domestic and international social media platforms - demonstrate superior performance of the proposed model in news detection tasks, achieving an accuracy rate

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Thymine-Hg2+-Thymine Strategy in Mof-Based Electrochemical Aptamer Sensor for Paes Detection

SSRN

引用

SSRN 2024年

作者： Liu, Xiaofang Diao, Zhan Liu, Huan Wang, Qun Lei, Jincan Huo, Danqun Hou, Jingzhou Hou, Changjun Key Laboratory for Biorheological Science and Technology Ministry of Education Bioengineering College Chongqing University Chongqing400044 China Chongqing Engineering and Technology Research Center of Intelligent Rehabilitation and Eldercare Chongqing City Management College Chongqing401331 China Key Laboratory of Condiment Supervision Technology for State Market Regulation Chongqing Institute for Food and Drug Control Chongqing401121 China Chongqing Key Laboratory of Bio-perception & Intelligent Information Processing School of Microelectronics and Communication Engineering Chongqing University Chongqing400044 China

Phthalate acid esters (PAEs), known food contaminants, can lead to reproductive problems, respiratory diseases, childhood obesity, and neuropsychological disorders. This work aims to develop an electrochemical aptamer sensor for the detection of DAP and its analogs. To achieve specific recognition ability for PAEs, We have introduced the thymidine-Hg2+-thymidine strategy for the first time in an electrochemical aptamer sensor, incorporating metal-organic frameworks (MOFs) as the carrier for the aptamer. Additionally, nanoamplification technology was applied by immobilizing MXene on the electrode to intensify the sensor's current signal. In the presence of PAEs, these molecules specifically bound to the aptamers, resulting in competitive release of Hg2+ and aptamers connected to Hg2+, subsequently reducing current signaling. Under optimal experimental conditions, the designed sensor achieved a linear range from 1.65×10-5 mg/mL to 3×10-2 mg/ml and an extremely low detection limit of 8.94×10-6 mg/mL. Furthermore, it exhibited good reliability and practicality in detecting actual samples. © 2024, The Authors. All rights reserved.

关键词： Esters

来源：评论

学校读者我要写书评

暂无评论

Salient Object Detection in Traffic Scene through the TSOD10K Dataset

arXiv

引用

arXiv 2025年

作者： Qiu, Yu Sun, Yuhang Mei, Jie Xiao, Lin Xu, Jing College of Information Science and Engineering Hunan Normal University Changsha410081 China National Key Laboratory of Intelligent Tracking and Forecasting for Infectious Diseases College of Artificial Intelligence Nankai University Tianjin300350 China National Engineering Research Center of Robot Visual Perception and Control Technology School of Robotics Hunan University Changsha410082 China

Traffic Salient Object Detection (TSOD) aims to segment the objects critical to driving safety by combining semantic (e.g., collision risks) and visual saliency. Unlike SOD in natural scene images (NSI-SOD), which prioritizes visually distinctive regions, TSOD emphasizes the objects that demand immediate driver attention due to their semantic impact, even with low visual contrast. This dual criterion, i.e., bridging perception and contextual risk, re-defines saliency for autonomous and assisted driving systems. To address the lack of task-specific benchmarks, we collect the first large-scale TSOD dataset with pixel-wise saliency annotations, named TSOD10K. TSOD10K covers the diverse object categories in various real-world traffic scenes under various challenging weather/illumination variations (e.g., fog, snowstorms, low-contrast, and low-light). Methodologically, we propose a Mamba-based TSOD model, termed Tramba. Considering the challenge of distinguishing inconspicuous visual information from complex traffic backgrounds, Tramba introduces a novel Dual-Frequency Visual State Space module equipped with shifted window partitioning and dilated scanning to enhance the perception of fine details and global structure by hierarchically decomposing high/low-frequency components. To emphasize critical regions in traffic scenes, we propose a traffic-oriented Helix 2D-Selective-Scan (Helix-SS2D) mechanism that injects driving attention priors while effectively capturing global multi-direction spatial dependencies. We establish a comprehensive benchmark by evaluating Tramba and 22 existing NSI-SOD models on TSOD10K, demonstrating Tramba’s superiority. Our research establishes the first foundation for safety-aware saliency analysis in intelligent transportation systems. Copyright © 2025, The Authors. All rights reserved.

关键词： Risk perception

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：