检索结果-内蒙古大学图书馆

17th ieee International symposium on Applied computational intelligence and Informatics, SACI 2023

作者： Al-Shouha, Modafar Szucs, Gabor Budapest University of Technology and Economics Faculty of Electrical Engineering and Informatics Department of Telecommunications and Media Informatics Műegyetem rkp. 3 BudapestH-1111 Hungary

ISBN: (纸本)9798350321104

The rapid advancement in Deep Learning (DL) proposes viable solutions to various real-world problems. However, deploying DL-based models in some applications is hindered by their black-box nature and the inability to explain them. This has pushed Explainable Artificial intelligence (XAI) research toward DL-based models, aiming to increase the trust by reducing their opacity. Although many XAI algorithms were proposed earlier, they lack the ability to explain certain tasks, i.e. image captioning (IC). This is caused by the IC task nature, e.g. the presence of multiple objects from the same category in the captioned image. In this paper we propose and investigate an XAI approach for this particular task. Additionally, we provide a method to evaluate XAI algorithms performance in the domain1.1Our code is available at https://***/modafarshouha/PIC-XAI. © 2023 ieee.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Performance Optimization of U-Net Network in Segmentation of Nerves image Dataset

Performance Optimization of U-Net Network in Segmentation of...

引用

2025 International Conference on Multi-Agent Systems for Collaborative intelligence, ICMSCI 2025

作者： Sreeja, G Veera Nagaraju, M. Theja, C.P. Sai Agarwal, Vikas Telangana Hyderabad India

ISBN: (纸本)9798331509828

This research explores the performance and optimization of U-Net architectures for segmenting nerve images. U-Nets are prevalent in medical image analysis due to their effectiveness in capturing detailed features and spatial relationships. Our focus is on enhancing U-Net models by fine-tuning hyperparameters, applying data augmentation techniques, and utilizing advanced loss functions to boost segmentation accuracy and computational efficiency. The study compares various U-Net variants, evaluating their effectiveness in preserving the delicate structures of nerve tissues. The results indicate that optimized U-Nets deliver high accuracy with reduced processing times, making them ideal for real-time medical applications. This research provides valuable insights into the practical use of U-Nets for nerve image segmentation and suggests avenues for future improvements. © 2025 ieee.

关键词： computational Modelling Data Augmentation Hyperparameter Tuning Medical image Analysis Nerve Disease Segmentation

来源：评论

学校读者我要写书评

暂无评论

CMUNEXT: AN EFFICIENT MEDICAL image SEGMENTATION NETWORK BASED ON LARGE KERNEL AND SKIP FUSION 21

CMUNEXT: AN EFFICIENT MEDICAL IMAGE SEGMENTATION NETWORK BAS...

引用

21st ieee International symposium on Biomedical Imaging (ISBI)

作者： Tang, Fenghe Ding, Jianrui Quan, Quan Wang, Lingtao Ning, Chunping Zhou, S. Kevin Univ Sci & Technol China Sch Biomed Engn Div Life Sci & Med Hefei 230026 Anhui Peoples R China Univ Sci & Technol China Suzhou Inst Adv Res Suzhou 215123 Jiangsu Peoples R China Harbin Inst Technol Sch Comp Sci & Technol Harbin Peoples R China Chinese Acad Sci Inst Comp Technol Beijing Peoples R China Qingdao Univ Affiliated Hosp Dept Ultrasound Qingdao Peoples R China

ISBN: (纸本)9798350313345;9798350313338

The u-shaped architecture has emerged as a crucial paradigm in the design of medical image segmentation networks. However, due to the inherent local limitations of convolution, a fully convolutional segmentation network with u-shaped architecture struggles to effectively extract global context information, which is vital for the precise localization of lesions. While hybrid architectures combining CNN and Transformer can address these issues, their applications are limited due to the computational resource. In addition, the inductive bias of convolution in lightweight networks adeptly fits the scarce medical data, which is lacking in the Transformer based network. To extract global context information while taking advantage of the inductive bias, we propose CMUNeXt, an efficient lightweight segmentation network, which leverages large kernel and inverted bottleneck design to thoroughly mix distant spatial and location information, efficiently extracting global context information. We also introduce the Skip-fusion block, designed to enable smooth skip-connections and ensure ample feature fusion. Experimental results demonstrate that CMUNeXt outperforms existing heavyweight and lightweight medical image segmentation networks, while offering a faster inference speed, lighter weights, and a reduced computational cost. The code is available at https://***/FengheTan9/CMUNeXt.

关键词： Medical image segmentation Light-weight network Large Kernel Skip-fusion

来源：评论

学校读者我要写书评

暂无评论

Comparative study of data augmentation approaches for improving medical image classification

Comparative study of data augmentation approaches for improv...

引用

International Conference on computational Science and computational intelligence (CSCI)

作者： Rais, Khadija Amroune, Mohamed Haouam, Mohamed Yassine Bendib, Issam Echahid Cheikh Larbi Tebessi Univ Lab Math Informat & Syst LAMIS Tebessa 12002 Algeria

ISBN: (纸本)9798350361513;9798350372304

In recent years, data augmentation has advanced to the point where it no longer relies on traditional photometric or geometric image processing techniques, such as rotation, scale, and filtering. Instead, it has turned to deep learning-based approaches like variational autoencoders (VAEs) and generative adversarial networks (GANs). This shift has made data augmentation a crucial discipline for improving artificial intelligence models, especially in fields like medicine where collecting labeled data is difficult and expensive. This study focuses on using data augmentation approaches to enhance convolutional neural networks (CNNs) classification accuracy. These approaches include geometric modifications, color space transformations, and generative techniques, and they are employed to augment the BraTS20 (Brain Tumor Segmentation 2020) dataset. After comparing the accuracy of CNNs with and without the augmented dataset, we found that DCGANs and AttnGAN achieved the highest accuracy among all techniques. They generated images that closely resembled those in BraTS20 compared to other methods.

关键词： Data augmentation classic techniques generative techniques Medical image classification Brain tumor

来源：评论

学校读者我要写书评

暂无评论

NOVEL OCT MOSAICKING PIPELINE WITH FEATURE- AND PIXEL-BASED REGISTRATION 21

NOVEL OCT MOSAICKING PIPELINE WITH FEATURE- AND PIXEL-BASED ...

引用

21st ieee International symposium on Biomedical Imaging (ISBI)

作者： Wang, Jiacheng Li, Hao Hu, Dewei Tao, Yuankai K. Oguz, Ipek Vanderbilt Univ Nashville TN 37235 USA

ISBN: (纸本)9798350313345;9798350313338

High-resolution Optical Coherence Tomography (OCT) images are crucial for ophthalmology studies but are limited by their relatively narrow field of view (FoV). image mosaicking is a technique for aligning multiple overlapping images to obtain a larger FoV. Current mosaicking pipelines often struggle with substantial noise and considerable displacement between the input sub-fields. In this paper, we propose a versatile pipeline for stitching multi-view OCT/OCTA en face projection images. Our method combines the strengths of learning-based feature matching and robust pixel-based registration to align multiple images effectively. Furthermore, we advance the application of a trained foundational model, Segment Anything Model (SAM), to validate mosaicking results in an unsupervised manner. The efficacy of our pipeline is validated using an in-house dataset and a large public dataset, where our method shows superior performance in terms of both accuracy and computational efficiency. We also made our evaluation tool for image mosaicking and the corresponding pipeline publicly available at https://***/MedICL-VU/OCT-mosaicking.

关键词： image mosaicking OCT Feature matching image registration Foundation model

来源：评论

学校读者我要写书评

暂无评论

Multi-Scale Fusion and Decomposition Network for Single image Deraining

引用

ieee Transactions on image processing 2024年 33卷 191-204页

作者： Wang, Qiong Jiang, Kui Wang, Zheng Ren, Wenqi Zhang, Jianhui Lin, Chia-Wen Wuhan University National Engineering Research Center for Multimedia Software School of Computer Science Wuhan430072 China Sun Yat-sen University School of Cyber Science and Technology Guangzhou510275 China Hangzhou Dianzi University College of Computer Science and Technology Hangzhou310018 China National Tsing Hua University Department of Electrical Engineering The Institute of Communications Engineering Hsinchu30013 Taiwan

Convolutional neural networks (CNNs) and self-attention (SA) have demonstrated remarkable success in low-level vision tasks, such as image super-resolution, deraining, and dehazing. The former excels in acquiring local connections with translation equivariance, while the latter is better at capturing long-range dependencies. However, both CNNs and Transformers suffer from individual limitations, such as limited receptive field and weak diversity representation of CNNs during low efficiency and weak local relation learning of SA. To this end, we propose a multi-scale fusion and decomposition network (MFDNet) for rain perturbation removal, which unifies the merits of these two architectures while maintaining both effectiveness and efficiency. To achieve the decomposition and association of rain and rain-free features, we introduce an asymmetrical scheme designed as a dual-path mutual representation network that enables iterative refinement. Additionally, we incorporate high-efficiency convolutions throughout the network and use resolution rescaling to balance computational complexity with performance. Comprehensive evaluations show that the proposed approach outperforms most of the latest SOTA deraining methods and is versatile and robust in various image restoration tasks, including underwater image enhancement, image dehazing, and low-light image enhancement. The source codes and pretrained models are available at https://***/qwangg/MFDNet. © 1992-2012 ieee.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Crack Detection with Conditional Diffusion Model 8

Crack Detection with Conditional Diffusion Model

引用

8th International Artificial intelligence and Data processing symposium, IDAP 2024

作者： Bostancioglu, Onur Karacan, Levent Iskenderun Technical University KocDigital Çözümler A.Ş Hatay Turkey Gaziantep University Gaziantep Turkey

ISBN: (纸本)9798331531492

Detecting cracks in surface images is crucial in engineering and materials science. Although learning-based methods have shown success in this area, accurately determining the shape of cracks remains a challenge. To address this issue, we introduce the first conditional diffusion model in crack detection. Our experiments show that this model can accurately detect crack shapes and achieves the highest precision score. © 2024 ieee.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

Investigating Effect of Optimization Algorithms on Chaos Based image Encryption 8

Investigating Effect of Optimization Algorithms on Chaos Bas...

引用

8th International Artificial intelligence and Data processing symposium, IDAP 2024

作者： Ince, Cemile Ince, Kenan İnönü University Computer Engineering Dept Malatya Turkey İnönü University Software Engineering Dept Malatya Turkey

ISBN: (纸本)9798331531492

image encryption is a specialized subfield of cryptography, distinguished by the high correlation between neighboring pixels, which necessitates unique approaches to disrupt this correlation. While all encryption approaches fundamentally involve confusion and diffusion operators, the specific implementations of these operators can vary significantly. In recent years, approaches that combine chaos theory with optimization techniques, which aim to find the optimal solution among numerous possibilities, have become prevalent. These studies, which aim to enhance the security and efficiency of the encryption process, such as by selecting the initial values of chaotic maps in regions with high chaotic behavior, show promise. This study examines the most commonly applied optimization methods in the field of cryptography within the context of image encryption, providing a comprehensive overview. Additionally, it empirically demonstrates that the optimization techniques employed in image encryption studies improve encryption quality. © 2024 ieee.

关键词： Encryption algorithms

来源：评论

学校读者我要写书评

暂无评论

Split Edge-Cloud Neural Networks for Better Adversarial Robustness

引用

ieee ACCESS 2024年 12卷 158854-158865页

作者： Douch, Salmane Abid, Mohamed Riduan Zine-Dine, Khalid Bouzidi, Driss Benhaddou, Driss Mohammed V Univ Rabat Natl Sch Comp Sci & Syst Anal ENSIAS Rabat 30050 Morocco Columbus State Univ TSYS Sch Comp Sci Columbus GA 31907 USA Mohammed V Univ Rabat Fac Sci FSR Rabat 30050 Morocco Alfaisal Univ Coll Engn Riyadh 11533 Saudi Arabia

Cloud computing is a critical component in the success of 5G and 6G networks, particularly given the computation-intensive nature of emerging applications. Despite all it advantages, cloud computing faces limitations in meeting the strict latency and bandwidth requirements of applications such as eHealth and automotive systems. To overcome these limitations, edge computing has emerged as a novel paradigm that bring computation closer to the user. Moreover, intelligent tasks such as deep learning ones demand more memory and processing power than edge devices can handle. To address these challenges, methods like quantization, pruning, and distributed inference have been proposed. Similarly, this paper study a promising approach for running deep learning models at the edge: split neural networks (SNN). SNNs feature a neural network architecture with multiple early exit points, allowing the model to make confident decisions at earlier layers without processing the entire network. This not only reduces memory and computational demands but it also makes SNNs well-suited for edge computing applications. As the use of SNNs expands, ensuring their safety-particularly their robustness to perturbations-becomes crucial for deployment in safety-critical scenarios. This paper presents the first in-depth study on the robustness of split Edge Cloud neural networks. We review state-of-the-art robustness certification techniques and evaluate SNN robustness using the auto_LiRPA and Auto Attack libraries, comparing them to standard neural networks. Our results demonstrate that SNNs reduce average inference time by 75'% and certify 4 to 10 times more images as robust, while improving overall robustness accuracy by 1% to 10%.

关键词： Robustness Edge computing Perturbation methods computational modeling Cloud computing Certification Biological neural networks Quantization (signal) image edge detection Deep learning Adversarial attacks cloud computing edge computing edge intelligence robustness certification split neural networks

来源：评论

学校读者我要写书评

暂无评论

VISUAL QUESTION ANSWERING FOR WISHART H-ALPHA CLASSIFICATION OF POLARIMETRIC SAR imageS

VISUAL QUESTION ANSWERING FOR WISHART H-ALPHA CLASSIFICATION...

引用

ieee International Geoscience and Remote Sensing symposium (IGARSS)

作者： Aghababaei, Hossein Stein, Alfred Univ Twente Fac Geoinformat Sci & Earth Observat Enschede Netherlands

ISBN: (纸本)9798350360332;9798350360325

Polarimetric Synthetic Aperture Radar (PolSAR) images offer a rich repository of information, crucial for diverse applications ranging from classification to target identification. In the domain of PolSAR image classification, the Wishart classifier emerged as a prominent and widely employed technique. This classifier is often used to articulate the properties of polarimetric scattering types in images, providing valuable insight into various types of targets. With the growing interest in multidisciplinary Artificial intelligence (AI) research, especially in computer vision and Natural Language processing (NLP), our goal is to integrate this enthusiasm into polarimetric image analysis. We propose extending the Wishart classifier framework to include a free-form and open-ended Visual Question Answering (VQA) model. This model is designed to answer natural language questions related to PolSAR images, covering pixel details and scattering patterns. The objective is to provide accurate natural language responses that reflect real-world scenarios, such as assisting the visually impaired. Both questions and answers in this context are intentionally left open-ended to capture the complexity of inquiries in the polarimetric SAR images domain.

关键词： Polarimetric SAR image Classification Visual Question Answering Deep Learning Natural Language Remote Sensing

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：