检索结果-内蒙古大学图书馆

RepVGG-MEM: A Lightweight Model for Garbage Classification Achieving a Balance Between Accuracy and Speed

ieee ACCESS 2025年 13卷 36451-36469页

作者： Si, Qiuxin Han, Sang Ik Zibo Vocat Inst Sch Artificial Intelligence & Big Data Zibo 255000 Shandong Peoples R China Semyung Univ Dept Informat & Commun Jecheon Si 27136 South Korea Semyung Univ Sch Smart IT Jecheon Si 27136 South Korea

Currently, existing garbage image classification models predominantly operate on low-end devices and encounter significant challenges, including limitations in computing resources, storage capacity, and classification accuracy. This paper proposes an improved lightweight model, RepVGG-MEM, specifically designed to address the resource constraints of low-end devices. The backbone of this model is derived from the lightweight RepVGG architecture, augmented by the integration of a multi-scale convolutional attention module to enhance high-quality feature extraction. Experimental results demonstrate that the RepVGG-MEM model outperforms its counterparts, achieving an accuracy of 93.26%, with a parameter count of 7.2 million and a floating-point operations (FLOPs) of 1.41 billion. This performance reflects a commendable balance between accuracy and processing speed. Furthermore, the model's redundancy is minimized through pruning techniques, which significantly reduce both complexity and computational overhead. The optimal pruned version of the model is designated as RepVGG-MEM5. In this iteration, the parameter count is further reduced to 1.2 million and the FLOPs decrease to 0.55 billion, with a minor accuracy decline of only 1.17%. These findings indicate that it is possible to significantly reduce the model's parameters and FLOPs without a substantial loss in accuracy, thereby enhancing the overall performance of the model while achieving an optimal balance of accuracy and speed. As a result, this research contributes to the development of an efficient and lightweight convolutional neural network model for garbage classification.

关键词： computational modeling Convolution Accuracy Kernel Biological system modeling Feature extraction Sorting image classification Data models Adaptation models Garbage image classification RepVGG multi-scale convolutional attention module pruning

来源：评论

学校读者我要写书评

暂无评论

Exploring Generative Adversarial Networks: Comparative Analysis of Facial image Synthesis and the Extension of Creative Capacities in Artificial intelligence

引用

ieee ACCESS 2025年 13卷 19588-19597页

作者： Eglynas, Tomas Lizdenis, Dovydas Raudys, Aistis Jakovlev, Sergej Voznak, Miroslav Klaipeda Univ Marine Res Inst LT-92294 Klaipeda Lithuania Vilnius Univ Fac Math & Informat LT-03225 Vilnius Lithuania VSB Tech Univ Ostrava Telecommun Dept Ostrava 70833 Czech Republic

Neural networks have become foundational in modern technology, driving advancements across diverse domains such as medicine, law enforcement, and information technology. By enabling algorithms to learn from data and perform tasks autonomously, they eliminate the need for explicit programming. A significant challenge in this field is replicating the uniquely human capacity for creativity-envisioning and realizing novel concepts and tangible creations. Generative Adversarial Networks (GANs), a leading approach in this effort, are especially notable for synthesizing realistic human facial images. Despite the success of GANs, comprehensive comparative studies of face-generating GAN methodologies are limited. This paper addresses this gap by analyzing the scope and capabilities of facial generation, detailing the principles of the original GAN framework, and reviewing prominent GAN variants specifically designed for facial synthesis. Through performance evaluations and fidelity analysis of generated images, this study contributes to a deeper understanding of GAN potential in advancing artificial intelligence creativity through performance evaluations and fidelity analysis of generated images.

关键词： Generative adversarial networks Generators Training Faces Noise Synthetic data Data models Loss measurement Decoding computational modeling Human image synthesis image processing computer graphics visualization photorealism

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Model for Fast and Detail-Preserving Human Segmentation in Video

引用

ieee TRANSACTIONS ON EMERGING TOPICS IN computational intelligence 2025年

作者： Zhang, Xiaoyao Yao, Lan Zeng, Feng Cent South Univ Sch Comp Scienceand Engn Changsha 410017 Peoples R China Hunan Univ Sch Math Changsha 410012 Peoples R China

With the rapid development of artificial intelligence, human segmentation in video is becoming increasingly important in the field of computer vision. However, existing segmentation models suffer from inaccurate segmentation, slow processing, and large model size that limit their deployment on resource-constrained devices. To this end, we propose a lightweight model called Efficient Memory Aggregation U-shaped Network (EMAUnet) for human segmentation in video, which is based on a traditional U-shaped network and attention mechanism. In EMAUnet, memory modules are combined with segmentation modules, enabling end-to-end learning of semantic extraction patterns for human images. Mobile-inverse bottleneck convolution (MBConv) is used as the network backbone that has relatively few parameters and computational complexity. Inverted sub-pixel down-sampling (ISP) is proposed to minimize information loss and achieve detail-preserving of segmentation. Coordinate attention (CA) is adopted to precisely locate the portrait area. Moreover, bidirectional memory update (BMU) and memory update trigger (MUT) are proposed to improve memory resource utilization and reduce unnecessary computation. Experimental results show that, compared with the classic model ISNet, EMAUnet has the mIoU, FPS and pixel accuracy increased by 2.3%, 25.0% and 1.8%, respectively, while the amount of parameters and the size of model decreased by 33.3% and 45.7%, respectively.

关键词： image segmentation computational modeling Accuracy Semantics Object segmentation Streaming media Real-time systems Memory management Convolution Training Video human segmentation lightweight model U-shaped network attention mechanism

来源：评论

学校读者我要写书评

暂无评论

IRAU-Net: Inception Residual Attention U-Net in Adversarial Network for Cardiac MRI Segmentation

引用

ieee JOURNAL OF SELECTED TOPICS IN SIGNAL processing 2025年第1期19卷 260-272页

作者： Rostami, Maryam Talebi Motamedi, Seyed Ahmad Amirkabir Univ Technol Elect Engn Dept Tehran *** Iran

Due to the significant advancements in medical imaging, the application of artificial intelligence for early disease diagnosis has greatly contributed to reducing mortality caused by heart diseases. Cardiac MRI segmentation into the left ventricle (LV), right ventricle (RV) and myocardium (MYO) is a key step towards the early treatment and diagnosis of heart diseases. The proposed method consists of two main components: a localization part to extract the heart organ from other parts of the image, and a GAN model for segmentation. In the generator part of the generative adversarial network (GAN) model, a customized U-Net network is employed to segment the cardiac image. A combination of inception and residual blocks and utilizing an attention mechanism yield an improved version of U-Net. On the other hand, the discriminator part of the GAN model is designed to accurately distinguish between the ground truth image and the segmented image generated by the generator part. Our method is evaluated on two cardiac MRI datasets. The evaluation results on the ACDC 2017 challenge dataset show mean Dice scores of 0.947 for LV, 0.919 for RV, and 0.907 for MYO in cardiac MRI segmentation. The experimental results highlight that our proposed IRAU-Net method outperforms other state-of-the-art methods in terms of accuracy while significantly reducing computational costs.

关键词： image segmentation Generators Generative adversarial networks Heart Accuracy Magnetic resonance imaging Location awareness Detectors Computer architecture Semantic segmentation Attention mechanism cardiac segmentation deep learning generative adversarial network medical image segmentation U-Net

来源：评论

学校读者我要写书评

暂无评论

Generalized Task-Driven Medical image Quality Enhancement With Gradient Promotion

引用

ieee TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE intelligence 2025年第4期47卷 2785-2798页

作者： Zhang, Dong Cheng, Kwang-Ting Hong Kong Univ Sci & Technol Dept Elect & Comp Engn Hong Kong Peoples R China

Thanks to the recent achievements in task-driven image quality enhancement (IQE) models like ESTR (Liu et al. 2023), the image enhancement model and the visual recognition model can mutually enhance each other's quantitation while producing high-quality processed images that are perceivable by our human vision systems. However, existing task-driven IQE models tend to overlook an underlying fact-different levels of vision tasks have varying and sometimes conflicting requirements of image features. To address this problem, this paper proposes a generalized gradient promotion (GradProm) training strategy for task-driven IQE of medical images. Specifically, we partition a task-driven IQE system into two sub-models, i.e., a mainstream model for image enhancement and an auxiliary model for visual recognition. During training, GradProm updates only parameters of the image enhancement model using gradients of the visual recognition model and the image enhancement model, but only when gradients of these two sub-models are aligned in the same direction, which is measured by their cosine similarity. In case gradients of these two sub-models are not in the same direction, GradProm only uses the gradient of the image enhancement model to update its parameters. Theoretically, we have proved that the optimization direction of the image enhancement model will not be biased by the auxiliary visual recognition model under the implementation of GradProm. Empirically, extensive experimental results on four public yet challenging medical image datasets demonstrated the superior performance of GradProm over existing state-of-the-art methods.

关键词： Medical diagnostic imaging Visualization image recognition image enhancement image quality Training Optimization Noise reduction IP networks computational modeling image quality enhancement medical image processing task-auxiliary learning multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Exoplanet Imaging via Differentiable Rendering

IEEE TRANSACTIONS ON COMPUTATIONAL IMAGING

引用

ieee TRANSACTIONS ON computational IMAGING 2025年 11卷 36-51页

作者： Feng, Brandon Y. Ferrer-Chavez, Rodrigo Levis, Aviad Wang, Jason J. Bouman, Katherine L. Freeman, William T. MIT Comp Sci & Artificial Intelligence Lab Cambridge MA 02139 USA Northwestern Univ Dept Phys & Astron Evanston IL 60208 USA Univ Toronto Dept Comp Sci Toronto ON M5S 1A1 Canada CALTECH Dept Comp & Math Sci Pasadena CA 91125 USA CALTECH Dept Elect Engn & Astron Pasadena CA 91125 USA

Direct imaging of exoplanets is crucial for advancing our understanding of planetary systems beyond our solar system, but it faces significant challenges due to the high contrast between host stars and their planets. Wavefront aberrations introduce speckles in the telescope science images, which are patterns of diffracted starlight that can mimic the appearance of planets, complicating the detection of faint exoplanet signals. Traditional post-processing methods, operating primarily in the image intensity domain, do not integrate wavefront sensing data. These data, measured mainly for adaptive optics corrections, have been overlooked as a potential resource for post-processing, partly due to the challenge of the evolving nature of wavefront aberrations. In this paper, we present a differentiable rendering approach that leverages these wavefront sensing data to improve exoplanet detection. Our differentiable renderer models wave-based light propagation through a coronagraphic telescope system, allowing gradient-based optimization to significantly improve starlight subtraction and increase sensitivity to faint exoplanets. Simulation experiments based on the James Webb Space Telescope configuration demonstrate the effectiveness of our approach, achieving substantial improvements in contrast and planet detection limits. Our results showcase how the computational advancements enabled by differentiable rendering can revitalize previously underexploited wavefront data, opening new avenues for enhancing exoplanet imaging and characterization.

关键词： Planets Imaging Stars Exoplanet Telescopes Rendering (computer graphics) Extraterrestrial measurements Optical imaging Optimization Optical diffraction Exoplanet imaging high-contrast imaging differentiable rendering wavefront aberration estimation

来源：评论

学校读者我要写书评

暂无评论

Generative Adversarial Network Based image-Scaling Attack and Defense Modeling

引用

ieee TRANSACTIONS ON EMERGING TOPICS IN computational intelligence 2025年第1期9卷 861-873页

作者： Li, Junjian Chen, Honglong Li, Zhe Zhang, Anqing Wang, Xiaomeng Wang, Xingang Xia, Feng China Univ Petr East China Coll Control Sci & Engn Qingdao 266580 Peoples R China Radboud Univ Nijmegen Fac Sci NL-6500 HC Nijmegen Netherlands Qilu Univ Technol Shandong Acad Sci Shandong Comp Sci Ctr Key Lab Comp Power Network & Informat SecurMinist Jinan 250353 Peoples R China RMIT Univ Sch Comp Technol Melbourne Vic 3000 Australia

Deep learning has achieved remarkable performance in computer vision applications, and gradually becomes one of the mainstream technologies. image scaling, as an indispensable data pre-processing procedure for most of the computer vision applications, is implemented to resize the unmatched data to fit the input sizes of deep learning models. However, such kind of data pre-processing is vulnerable to be exploited to launch an attack called image-scaling attack, which can generate attack images presenting entirely different appearances after scaling. In this paper, we develop the attack model based on generative adversarial network (GAN) to fuse the source and target images, called Attack-GAN, which guarantees the generated attack images are imperceptible as well as being scaled to the target images. Experiments show that the superiority of Attack-GAN over the optimization method is that it can generate attack images with better camouflage and achieve higher attack success rates as well as accelerate the generation of attack images. Furthermore, we propose the Defense-GAN to learn and approximate the distribution of unperturbed images from attack images, which aims to eliminate and compensate the adversarial pixels. Based on different capabilities, customized defense strategies for different defenders are developed to resist the image-scaling attack. Experimental results manifest the proposed defense strategies are effective against the image-scaling attack, i.e., the generator of Defense-GAN can recover the attack images and retain the original 'semantics'.

关键词： Generative adversarial networks Generators computational modeling Training Optimization Deep learning Data models image-scaling attack generative adversarial network adversarial examples transfer learning

来源：评论

学校读者我要写书评

暂无评论

Single-Node Power Demand During AI Training: Measurements on an 8-GPU NVIDIA H100 System

引用

ieee ACCESS 2025年 13卷 61740-61747页

作者： Latif, Imran Newkirk, Alex C. Carbone, Matthew R. Munir, Arslan Lin, Yuewei Koomey, Jonathan Yu, Xi Dong, Zhihua Brookhaven Natl Lab Comp & Data Sci Directorate Upton NY 11973 USA Lawrence Berkeley Natl Lab Bldg Technol & Urban Syst Berkeley CA 94720 USA Florida Atlantic Univ Dept Elect Engn & Comp Sci Boca Raton FL 33431 USA Koomey Analyt San Francisco CA 94011 USA

The expansion of artificial intelligence (AI) applications has driven substantial investment in computational infrastructure, especially by cloud computing providers. Quantifying the energy footprint of this infrastructure requires models parameterized by the power demand of AI hardware during training. In this work, we measured the instantaneous power draw of an 8-GPU NVIDIA H100 HGX node during the training of open-source image classifier (ResNet) and large-language models (Llama2-13b). We characterize power demand for a single node configuration, providing foundational data for future multi-node studies. The maximum observed power draw was approximately 8.4 kW, 18% lower than the manufacturer-rated 10.2 kW, even with GPUs near full utilization. Holding model architecture constant, increasing batch size from 512 to 4096 images for ResNet reduced total training energy consumption by a factor of 4. These findings can inform capacity planning for data center operators and energy use estimates by researchers. Future work will investigate the impact of cooling technology and carbon-aware scheduling on AI workload energy consumption.

关键词： Artificial intelligence Training Hardware Power measurement Graphics processing units Power demand Benchmark testing Visualization Stress computational modeling AI training sustainable computing GPU power measurements large language models

来源：评论

学校读者我要写书评

暂无评论

Mosaic-Mixed Attention-Based Unexpected Traffic Scene Classification

引用

ieee ACCESS 2025年 13卷 15712-15722页

作者： Lee, Sang-Hyun Lee, Soomok Yun, Ilsoo Ajou Univ Dept Data Network & AI Suwon 16499 South Korea Ajou Univ Dept Mobil Engn Suwon 16499 South Korea Ajou Univ Dept Transportat Syst Engn Suwon 16499 South Korea

With the rapid advancements in artificial intelligence and smart mobility technologies, traffic monitoring systems are evolving quickly. Among these systems, in-vehicle monitoring systems using Object Detection (OD) algorithms are gaining attention for identifying traffic participants in distress. However, current OD algorithms often underperform in complex or unexpected traffic scenarios, such as accidents. In this study, we propose a novel scene classification network that integrates OD with a cross-attention mechanism By leveraging spatial mosaic and mixed attention mechanisms, the network emphasizes spatial relationships and inter-channel correlations, significantly enhancing accuracy in identifying critical traffic events. The detailed evaluation demonstrates improved efficiency and accuracy, underscoring the potential of the system for future traffic incident classification.

关键词： Accidents Feature extraction Roads Classification algorithms Attention mechanisms Accuracy computational modeling Cameras Scene classification Convolutional neural networks Traffic monitoring scene classification attention network image mosaic processing

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

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：