检索结果-内蒙古大学图书馆

ieee International symposium on Circuits and Systems (ISCAS)

作者： Owada, Takumi Jin'no, Kenya Tokyo City Univ Fac Informat Technol Tokyo Japan

ISBN: (纸本)9798350330991;9798350331004

In this article, we propose the construction of a Convolutional Neural Network (CNN) model as a novel approach to detect fake images generated by diffusion models. However, constructing a CNN model to efficiently detect high-quality fake images may consume substantial computational resources, emphasizing the need for the development of a computationally inexpensive, lightweight model. At the crux of our research is the inquiry into whether a CNN can accurately detect counterfeit images produced by diffusion models using only partial image information. Further, to elucidate how this detection mechanism specifically functions, we delve deeply into the attention mechanism of the CNN. Experimental results indicate that the CNN places particular emphasis on edge information in its image analysis. This could potentially be key to effectively pinpointing specific features of counterfeit images generated by diffusion models.

关键词： CNN image classification Grad-CAM generative model

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Analysis of Transient Errors on Systolic Arrays 26

A Comprehensive Analysis of Transient Errors on Systolic Arr...

引用

26th International symposium on Design and Diagnostics of Electronic Circuits and Systems (DDECS)

作者： Vacca, Eleonora Azimi, Sarah Sterpone, Luca Politecn Torino Dipartimento Automat & Informat Turin Italy

ISBN: (纸本)9798350332773

In recent years, the growth of interest in adopting deep neural network techniques across various domains led to new architectures for supporting the required computational effort. Tensor processing Units (TPUs), which are based on a systolic array matrix multiplication unit (MMU), became widely popular thanks to their specific structure suitable for Artificial intelligence. This work investigates Single Event Transient (SET) effects on TPU's MMU. The analysis demonstrates the impact of SETs on the functionality of MMU when executing digital image processing filtering. The experimental results identify the static and dynamic SET sensitivity of TPU and depict meaningful information on the data dependency of the filters' kernel values.

关键词： AI Hardware Accelerator Single Event Transient Soft Errors Tensor processing Unit

来源：评论

学校读者我要写书评

暂无评论

Handwriting image Classification for Automated Diagnosis of Learning Disabilities: A Review on Deep Learning Models and Future Directions 19

Handwriting Image Classification for Automated Diagnosis of ...

引用

19th International Joint symposium on Artificial intelligence and Natural Language processing

作者： Sukiman, Safura Adeela Husin, Nor Azura Hamdan, Hazlina Murad, Masrah Azrifah Azmi Univ Teknologi MARA UiTM Johor Branch Coll Comp Informat & Math Segamat Campus Johor Baharu Malaysia Univ Putra Malaysia UPM Fac Comp Sci & Informat Technol Dept Comp Sci Serdang Selangor Malaysia

ISBN: (纸本)9798331509927;9798331509910

This study reviews deep learning models used in handwriting image classification for the automated diagnosis of learning disabilities. By addressing handwriting diversity and misclassification challenges, two models were highlighted: Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs). Literature was retrieved from major databases including ieee Xplore, Scopus, Web of Science (WoS), and Google Scholar, with studies on Parkinson's disease, tremor patients, and machine learning excluded. CNNs represent a more mature architecture focusing on convolutions, pooling, and activation function. Meanwhile, ViTs emerges as a promising alternative via its multi-head attention architecture. This review also compares the accuracy of both models, specifying the sources of handwriting images, as well as providing future directions relevant to the research field.

关键词： automated diagnosis deep learning handwriting classification learning disabilities

来源：评论

学校读者我要写书评

暂无评论

Content-decoupled Contrastive Learning-based Implicit Degradation Modeling for Blind image Super-Resolution Supplementary Material

引用

ieee Transactions on image processing 2025年 34卷 3013-3024页

作者： Yuan, Jiang Ma, Ji Wang, Bo Hu, Weiming North China Electric Power University School of Control and Computer Engineering Beijing102206 China CASIA State Key Laboratory of Multimodal Artificial Intelligence Systems Beijing100190 China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing100049 China ShanghaiTech University School of Information Science and Technology Shanghai201210 China

Implicit degradation modeling-based blind super-resolution (SR) has attracted more increasing attention in the community due to its excellent generalization to complex degradation scenarios and wide application range. How to extract more discriminative degradation representations and fully adapt them to specific image features is the key to this task. In this paper, we propose a new Content-decoupled Contrastive Learning-based blind image super-resolution (CdCL) framework following the typical blind SR pipeline. This framework introduces negative-free contrastive learning technique for the first time to model the implicit degradation representation, in which a new cyclic shift sampling strategy is designed to ensure decoupling between content features and degradation features from the data perspective, thereby improving the purity and discriminability of the learned implicit degradation space. In addition, we propose a detail-aware implicit degradation adapting module that can better adapt degradation representations to specific LR features by enhancing the basic adaptation unit's perception of image details, significantly reducing the overall SR model complexity. Extensive experiments on synthetic and real data show that our method achieves highly competitive quantitative and qualitative results in various degradation settings while obviously reducing parameters and computational costs, validating the feasibility of designing practical and lightweight blind SR tools. Codes and models will be available at https://***/Fieldhunter/CdCL. © 1992-2012 ieee.

关键词： Depression Anxiety disorders Face recognition Schizophrenia Visualization Training Data collection Uncertainty Facial features Emotion recognition

来源：评论

学校读者我要写书评

暂无评论

Context-Adaptive Deep Learning for Efficient image Parsing in Remote Sensing: An Automated Parameter Selection Approach

Context-Adaptive Deep Learning for Efficient Image Parsing i...

引用

2023 ieee symposium Series on computational intelligence, SSCI 2023

作者： Azam, Basim Verma, Brijesh Zhang, Mengjie Brisbane Australia Cdsai Victoria University of Wellington Wellignton New Zealand

ISBN: (纸本)9781665430654

image parsing is among the core tasks in the field of image processing and computer vision having wide-ranging applications in the areas of autonomous driving, image interpretation, medical analysis, and remote sensing. The modern techniques despite performing the labeling tasks accurately face several challenges. Among these challenges, the computation of contextual information and the selection of optimized parameters are of prime importance in pixel-wise segmentation tasks. We propose a novel context adaptive image parsing framework that utilizes a unique parameter selection strategy to produce final pixel labels. The automatic parameter selection minimizes the computational overhead, reduces the time complexity, and improves upon the segmentation labels produced. The proposed framework is evaluated on Wuhan Dense Labelling Dataset (WHDLD). In addition, a comprehensive comparison with state- of-the-art image segmentation techniques is presented. Finally, the analysis supporting the dominance of proposed architecture is presented in collation with existing techniques. © 2023 ieee.

关键词： deep learning image parsing neural networks remote sensing

来源：评论

学校读者我要写书评

暂无评论

image Recognition for Wildlife Conservation

Image Recognition for Wildlife Conservation

引用

2025 ieee International Conference on computational, Communication and Information Technology, ICCCIT 2025

作者： Charanya, P. Sridharan, S. Yuvan Shankar, S. Sriram, M. Yashvanth, K.P. Department of Artificial Intelligence and Machine Learning Coimbatore India Department of Artificial Intelligence and Data Science Coimbatore India

ISBN: (数字)9798331512965

ISBN: (纸本)9798331512965

This study adopts an empirical approach to evaluate the efficacy of image processing methods in conservation efforts for animals. The initial phase involves the collection of data from various sources within the natural ecosystem, including images captured through camera traps, drones, and other monitoring devices. This comprehensive and diverse dataset is then processed using advanced image processing Tools. A key strength of this methodology lies in its use of Machine Learning algorithms, particularly YOLO v8 (You Only Look Once) and Support Vector Machines (SVMs). These techniques are suitable for wildlife conservation because of their capacity to analyze complex visual data and detect patterns indicative of species presence, behavior, and habitat conditions. The paper outlines the application of YOLO v8 and SVMs to the collected image data, enabling accurate identification, real-time detection, and classification of wildlife species. This capability is essential for monitoring populations, tracking movements, and identifying potential threats, thus facilitating timely conservation actions. Furthermore, the research extends its evaluation by comparing the results of the Machine Learning approach with traditional ecological survey methods, offering a comprehensive assessment of the effectiveness of image processing in wildlife conservation. © 2025 ieee.

关键词： Invertebrates

来源：评论

学校读者我要写书评

暂无评论

Joint Spatial and Frequency Domain Learning for Lightweight Spectral image Demosaicing

引用

ieee TRANSACTIONS ON image processing 2025年 34卷 1119-1132页

作者： Wu, Fangfang Huang, Tao Xu, Junwei Cao, Xun Dong, Weisheng Dong, Le Shi, Guangming Xidian Univ Sch Comp Sci & Technol Xian 710071 Peoples R China Xidian Univ Sch Artificial Intelligence Xian 710071 Peoples R China Yealink Xiamen 361009 Peoples R China Nanjing Univ Sch Elect Sci & Engn Nanjing 210093 Peoples R China

Conventional spectral image demosaicing algorithms rely on pixels' spatial or spectral correlations for reconstruction. Due to the missing data in the multispectral filter array (MSFA), the estimation of spatial or spectral correlations is inaccurate, leading to poor reconstruction results, and these algorithms are time-consuming. Deep learning-based spectral image demosaicing methods directly learn the nonlinear mapping relationship between 2D spectral mosaic images and 3D multispectral images. However, these learning-based methods focused only on learning the mapping relationship in the spatial domain, but neglected valuable image information in the frequency domain, resulting in limited reconstruction quality. To address the above issues, this paper proposes a novel lightweight spectral image demosaicing method based on joint spatial and frequency domain information learning. First, a novel parameter-free spectral image initialization strategy based on the Fourier transform is proposed, which leads to better initialized spectral images and eases the difficulty of subsequent spectral image reconstruction. Furthermore, an efficient spatial-frequency transformer network is proposed, which jointly learns the spatial correlations and the frequency domain characteristics. Compared to existing learning-based spectral image demosaicing methods, the proposed method significantly reduces the number of model parameters and computational complexity. Extensive experiments on simulated and real-world data show that the proposed method notably outperforms existing spectral image demosaicing methods.

关键词： Frequency-domain analysis Imaging image reconstruction Transformers Feature extraction Correlation Fourier transforms Interpolation Three-dimensional displays Optimization Spectral image demosaicing multispectral filter array (MSFA) deep learning spatial domain frequency domain transformer

来源：评论

学校读者我要写书评

暂无评论

LYT-NET: Lightweight YUV Transformer-Based Network for Low-Light image Enhancement

引用

ieee SIGNAL processing LETTERS 2025年 32卷 2065-2069页

作者： Brateanu, Alexandru Balmez, Raul Avram, Adrian Orhei, Ciprian Ancuti, Cosmin Univ Manchester Manchester M13 9PL Lancs England Univ Politehn Timisoara Timisoara 300006 Romania West Univ Timisoara Timisoara 300223 Romania

This letter introduces LYT-Net, a novel lightweight transformer-based model for low-light image enhancement. LYT-Net consists of several layers and detachable blocks, including our novel blocks-Channel-Wise Denoiser (CWD) and Multi-Stage Squeeze & Excite Fusion (MSEF)-along with the traditional Transformer block, Multi-Headed Self-Attention (MHSA). In our method we adopt a dual-path approach, treating chrominance channels U and V and luminance channel $Y$ as separate entities to help the model better handle illumination adjustment and corruption restoration. Our comprehensive evaluation on established LLIE datasets demonstrates that, despite its low complexity, our model outperforms recent LLIE methods.

关键词： Transformers Feature extraction image enhancement image restoration Training image color analysis Complexity theory Head computational modeling Convolution Low-light image enhancement vision transformer deep learning

来源：评论

学校读者我要写书评

暂无评论

image Quality Assessment Based on Multi-Scale Representation and Shifting Transformer

引用

ieee ACCESS 2025年 13卷 24276-24286页

作者： Fu, Geng Wang, Ziyu Zhang, Cuijuan Qi, Zerong Hu, Mingzheng Fu, Shujun Zhang, Yunfeng Shandong Univ Finance & Econ Sch Comp & Artificial Intelligence Jinan 250014 Peoples R China Yidu Cent Hosp Weifang Dept Intervent Therapy Qingzhou Peoples R China Shandong Chengshi Elect Technol Co Ltd Jinan 250002 Peoples R China Shandong Univ Sch Math Jinan 250100 Peoples R China

In automatic control systems, sensors and cameras are often used to capture images of the environment or processes being monitored. The quality of these images is paramount as it directly affects the system's ability to accurately interpret and respond to the visual information. image Quality Assessment (IQA) is a crucial metric for intelligent control systems and computer vision tasks, such as surveillance, restoration, and fingerprint identification, significantly advancing algorithm development in these areas. Recently, transformer-based algorithms have excelled in computer vision, particularly in image classification, surpassing convolutional neural network (CNN) methods. To enhance IQA using transformers, we propose Swin-MIQT, a multi-scale spatial pooling transformer with shifted windows. As a no-reference (NR) IQA method, Swin-MIQT processes images at their original resolution without resizing or cropping, unlike standard vision transformers. By using shifted windows, we reduce computational load through efficient self-attention processing. Additionally, a spatial pyramid pooling layer captures diverse image quality information, improving IQA accuracy for distorted images. Comprehensive experiments show that Swin-MIQT achieves state-of-the-art performance on three synthetic distortion databases (LIVE, LIVE MD, TID2013) and competitive results on three authentic distortion databases (LIVE Challenge, KonIQ-10K, SPAQ). The outstanding performance demonstrates that Swin-MIQ possesses robust learning and generalization capabilities across all referenced distorted databases.

关键词： Databases Transformers image quality computational modeling Benchmark testing Distortion Spatial resolution Residual neural networks Predictive models Measurement image quality assessment multi-scale no-reference spatial pooling shifted window transformer

来源：评论

学校读者我要写书评

暂无评论

Hybrid-Smash: A Heterogeneous CPU-GPU Compression Library

引用

ieee ACCESS 2024年 12卷 32706-32723页

作者： Penaranda, Cristian Reano, Carlos Silla, Federico Univ Politecn Valencia DISCA Valencia 46022 Spain Univ Valencia Dept Informat ETSE Burjassot 46100 Spain

Compression algorithms are widely used to reduce data size and improve application performance. Nevertheless, data compression has a computational cost which can limit its use. GPUs could be leveraged to reduce compression time. However, existing GPU-based compression libraries expect data to compress in GPU memory, although it is usually stored in CPU memory. Additionally, setup time of GPUs could be a problem when compressing small data sizes. In this paper, we implement a new GPU-based compression library. Contrary to existing ones, our library uses data located in CPU memory. Performance results show that, for the same compression algorithms, GPUs are beneficial for larger data sizes whereas smaller data sizes are compressed faster using CPUs. Therefore, we enhance our proposal with Hybrid-Smash: a heterogeneous CPU-GPU compression library, which transparently uses CPU or GPU compression depending on data size, thus improving compression for any data size.

关键词： Graphics processing units image coding Compression algorithms Symbols Proposals Artificial intelligence Parallel processing Performance evaluation Central processing Unit Hybrid intelligent systems Data systems Size control Lossless compression parallel computing GPU

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：