检索结果-内蒙古大学图书馆

LW-CovidNet: Automatic covid-19 lung infection detection from chest X-ray images

IET image processing 2023年第2期17卷 362-374页

作者： Ahmed, Noor Tan, Xin Ma, Lizhuang Shanghai Jiao Tong Univ Sch Elect Informat & Elect Engn Dept Comp Sci & Engn Shanghai Peoples R China

Coronavirus Disease 2019 (Covid-19) overtook the worldwide in early 2020, placing the world's health in threat. Automated lung infection detection using Chest X-ray images has a ton of potential for enhancing the traditional covid-19 treatment strategy. However, there are several challenges to detect infected regions from Chest X-ray images, including significant variance in infected features similar spatial characteristics, multi-scale variations in texture shapes and sizes of infected regions. Moreover, high parameters with transfer learning are also a constraints to deploy deep convolutional neural network(CNN) models in real time environment. A novel covid-19 lightweight CNN(LW-CovidNet) method is proposed to automatically detect covid-19 infected regions from Chest X-ray images to address these challenges. In our proposed hybrid method of integrating Standard and Depth-wise Separable convolutions are used to aggregate the high level features and also compensate the information loss by increasing the Receptive Field of the model. The detection boundaries of disease regions representations are then enhanced via an Edge-Attention method by applying heatmaps for accurate detection of disease regions. Extensive experiments indicate that the proposed LW-CovidNet surpasses most cutting-edge detection methods and also contributes to the advancement of state-of-the-art performance. It is envisaged that with reliable accuracy, this method can be introduced for clinical practices in the future.

关键词： traditional covid-19 treatment strategy Biology and medical computing feature extraction diseases X-ray imaging image classification disease regions representations medical computing automated lung infection detection chest X-ray images X-rays and particle beams (medical uses) neural nets detection boundaries Chest X-ray images Optical, image and video signal processing Computer vision and image processing techniques convolutional neural nets X-ray techniques: radiography and computed tomography (biomedical imaging/measurement) covid-19 infected regions deep learning (artificial intelligence) image recognition epidemics edge detection Automatic covid-19 lung infection detection diagnostic radiography cutting-edge detection methods Patient diagnostic methods and instrumentation medical image processing infected features similar spatial characteristics lung

来源：评论

学校读者我要写书评

暂无评论

Hybrid Spatial-Spectral Autoencoder Models for Lossy Satellite image Compression

引用

JOURNAL OF AEROSPACE INFORMATION SYSTEMS 2025年第5期22卷 336-357页

作者： Badr, Mohamed Ahmed Elrewainy, Ahmed Fathy Elshafey, Mohamed Abdelmoneim Taha Mil Tech Coll Avion Engn Dept Elect Engn Branch Cairo 11765 Egypt Mil Tech Coll Comp Engn & Artificial Intelligence Dept Elect Engn Branch Cairo 11765 Egypt

The challenge of handling vast amounts of high-resolution satellite imagery is driven by onboard memory and bandwidth limitations. As spatial and spectral resolutions increase, image compression, particularly deep-learning-based methods, is essential to overcome these limitations. This paper presents hybrid autoencoder models that combine convolutional neural networks, long short-term memory networks, and attention mechanisms for spatial and spectral feature extraction. The proposed architectures, including sparse and variational autoencoder counterparts, form a comprehensive image compression framework with quantization, and various entropy coders are applied to the EuroSat dataset (RGB and multispectral). The experimental results show the models' superiority over the JPEG family and recent state-of-the-art methods, achieving up to 3.3, 1.4, and 0.6% improvements in the peak signal-to-noise ratio, structural similarity index, and multiscale structural similarity index, respectively. Moreover, performance analysis in terms of computational complexity, processing time, and memory usage highlights the efficiency of the proposed models. A case study conducted on a real scene from the Sentinel-2 satellite further validates the compatibility of the proposed models with modern artificial intelligence chipsets.

关键词： Convolutional neural Network Long Short-Term Memory Attention mechanisms Spatial Spectral Feature Extraction AI-Chipsets Compatibility Computational Complexity Performance Satellite image Compression Hybrid Deep Learning Models Earth Observation Satellite Technology Smart Satellites

来源：评论

学校读者我要写书评

暂无评论

FEUNet: a flexible and effective U-shaped network for image denoising

引用

signal image AND VIDEO processing 2023年第5期17卷 2545-2553页

作者： Wu, Wencong Lv, Guannan Liao, Shicheng Zhang, Yungang Yunnan Normal Univ Sch Informat Sci 798 Juxian St Kunming 650500 Yunnan Peoples R China

Over the recent years, deep convolutional neural networks based models have been absolutely attractive in image denoising field due to their favorable performance. However, many existing deep neural network based image denoising models lack flexibility for spatially variant or real-world noise, which restricts the application of these models in real denoising scenes. In this paper, we propose a flexible and effective U-shaped network (FEUNet), which is effective in a wide range of noise levels, and can deal with spatially variant noise. The adjustable noise level map is used as the input of the FEUNet to enhance its flexibility. The U-Net is utilized to enhance the effectiveness of the proposed model. Experimental results have verified that the proposed FEUNet can obtain competitive denoising performances on many denoising tasks compared with the state-of-the-art denoising methods, which makes the proposed FEUNet well suited for the practical image denoising tasks.

关键词： image denoising Deep neural networks U-Net Noise level map

来源：评论

学校读者我要写书评

暂无评论

Diffusion Models for Audio Restoration

引用

IEEE signal processing MAGAZINE 2024年第6期41卷 72-84页

作者： Lemercier, Jean-Marie Richter, Julius Welker, Simon Moliner, Eloi Vaelimaeki, Vesa Gerkmann, Timo Univ Hamburg Hamburg Germany Aalto Univ Acoust Lab Espoo Finland Aalto Univ Espoo Finland

With the development of audio playback devices and fast data transmission, the demand for high sound quality is rising for both entertainment and communications. In this quest for better sound quality, challenges emerge from distortions and interferences originating at the recording side or caused by an imperfect transmission pipeline. To address this problem, audio restoration methods aim to recover clean sound signals from the corrupted input data. We present here audio restoration algorithms based on diffusion models, with a focus on speech enhancement and music restoration tasks. Traditional approaches, often grounded in handcrafted rules and statistical heuristics, have shaped our understanding of audio signals. In the past decades, there has been a notable shift toward data-driven methods that exploit the modeling capabilities of deep neural networks (DNNs). Deep generative models, and among them diffusion models, have emerged as powerful techniques for learning complex data distributions. However, relying solely on DNN-based learning approaches carries the risk of reducing interpretability, particularly when employing end-to-end models. Nonetheless, data-driven approaches allow more flexibility in comparison to statistical model-based frameworks, whose performance depends on distributional and statistical assumptions that can be difficult to guarantee. Here, we aim to show that diffusion models can combine the best of both worlds and offer the opportunity to design audio restoration algorithms with a good degree of interpretability and a remarkable performance in terms of sound quality. In this article, we review the use of diffusion models for audio restoration. We explain the diffusion formalism and its application to the conditional generation of clean audio signals. We believe that diffusion models open an exciting field of research with the potential to spawn new audio restoration algorithms that are natural-sounding and remain robust in difficult acoust

关键词： Inverse problems Acoustic noise signal processing algorithms stochastic processes Speech enhancement Diffusion models Robustness Trajectory Machine listening

来源：评论

学校读者我要写书评

暂无评论

GABIC: GRAPH-BASED ATTENTION BLOCK FOR image COMPRESSION 31

GABIC: GRAPH-BASED ATTENTION BLOCK FOR IMAGE COMPRESSION

引用

2024 International Conference on image processing

作者： Spadaro, Gabriele Presta, Alberto Tartaglione, Enzo Giraldo, Jhony H. Grangetto, Marco Fiandrotti, Attilio Univ Turin Turin Italy T elecom Paris Inst Polytech Paris LTCI Paris France

ISBN: (纸本)9798350349405;9798350349399

While standardized codecs like JPEG and HEVC-intra represent the industry standard in image compression, neural Learned image Compression (LIC) codecs represent a promising alternative. In detail, integrating attention mechanisms from Vision Transformers into LIC models has shown improved compression efficiency. However, extra efficiency often comes at the cost of aggregating redundant features. This work proposes a Graph-based Attention Block for image Compression (GABIC), a method to reduce feature redundancy based on a k-Nearest Neighbors enhanced attention mechanism. Our experiments show that GABIC outperforms comparable methods, particularly at high bit rates, enhancing compression performance.

关键词： Learned image Compression Graph neural Network Attention Mechanism Vision Transformer

来源：评论

学校读者我要写书评

暂无评论

An optimized facial emotion recognition architecture based on a deep convolutional neural network and genetic algorithm

引用

signal image AND VIDEO processing 2024年第2期18卷 1119-1129页

作者： Aghabeigi, Fereshteh Nazari, Sara Eraghi, Nafiseh Osati Islamic Azad Univ Dept Comp Engn & Informat Technol Arak Branch Arak Iran

Emotion recognition is one of the most interesting subjects in machine learning and computer vision fields, which is recognized by body language, speech, and face. Automatic emotion recognition is used in a variety of applications. In practice, recognizing human emotions with high accuracy is a challenging task. For this purpose, in this paper, we have recognized emotion from facial images using convolutional neural network architecture as one of the deep learning networks that used inception modules and dense blocks. The new proposed architecture is represented as GA-Dense-FaceliveNet, in which a genetic algorithm is expressed to tune the hyperparameters of the deep convolutional neural network. The proposed model is evaluated using three well-known datasets: CK + (extended Cohn-Kanade), JAFFE (Japanese Female Facial Expression), and KDEF (Karolinska Directed Emotional Faces). In the experiment, the accuracy of using CK + , JAFFE, and KDEF datasets is 99.96%, 98.92%, and 99.17%, respectively. The results demonstrate that the proposed method has higher performance compared to the state-of-the-art methods.

关键词： Deep learning Emotion recognition Facial emotion recognition Convolutional neural network Genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

Improving burn diagnosis in medical image retrieval from grafting burn samples using B-coefficients and the CLAHE algorithm

引用

BIOMEDICAL signal processing AND CONTROL 2025年 99卷

作者： Rangaiah, Pramod K. B. Kumar, B. P. Pradeep Augustine, Robin Uppsala Univ Dept Elect Engn Div Solid State Elect Microwaves Med Engn Grp Box 65 SE-75103 Uppsala Sweden Atria Inst Technol Dept Comp Sci & Design Bengaluru 560024 India

This study focuses on the vital difficulty of burn assessment in medical image retrieval from grafted burn specimens particularly in resource-constrained contexts where speedy and precise diagnoses are required. Our solution combines sophisticated machine learning techniques, namely an Artificial neural Network (ANN), with the Contrast Limited Adaptive Histogram Equalisation (CLAHE) algorithm in an image Reclamation system. The statistical assessments of kurtosis value (K-CLAHE=144.83) compared to the query image (K-query=131.17) indicate a distribution with more pronounced tails in the CLAHE image, enhancing specific image features. Additionally, increased skewness in the CLAHE image (S-CLAHE=5.92) suggests a shift toward higher intensity levels compared to the query image (S-query=4.47), further enhancing discernible image features. Through this incorporation, we carefully retain picture boundaries, boost local contrast, and minimize noise, hence enhancing burn diagnostic accuracy. Statistical analyses, such as kurtosis and skewness analysis, verify the improvements in visible picture aspects, offering significant insights into fundamental texture properties. We increase picture retrieval efficiency by using Bhattacharya coefficients and unique bin analysis, resulting in substantial enhancements in the retrieving score of matched images The ANN successfully differentiates between photos that require grafts and those that do not, providing a speedy and accurate diagnosis for acute burn injuries. This comprehensive technique greatly improves burn diagnosis, especially during emergencies, and shows promise for improving medical procedures. Our study helps to raise patient care standards in difficult medical situations by combining automated evaluation tools, powerful methods for image processing, and machine learning.

关键词： Artificial neural Network (ANN) image reclamation Burnt skin Medical imaging Bhattacharya coefficients Machine learning

来源：评论

学校读者我要写书评

暂无评论

CascadeMedSeg: integrating pyramid vision transformer with multi-scale fusion for precise medical image segmentation

引用

signal image AND VIDEO processing 2024年第12期18卷 9067-9079页

作者： Li, Junwei Sun, Shengfeng Li, Shijie Xia, Ruixue Henan Univ Sch Artificial Intelligence Zhengzhou 450046 Peoples R China Henan Univ Huaihe Hosp Kaifeng 475000 Peoples R China

Medical image segmentation (MIS) is a key technique in computer-aided diagnosis. With the development of deep learning, especially convolutional neural networks, the performance of MIS has been significantly improved, however, some mainstream convolution-based methods still suffer from inaccurate target boundaries and imprecise segmentation results. At the same time, transformer-based methods have gradually achieved better segmentation results. To overcome the challenges of traditional methods, an accurate MIS model (CascadeMedSeg) is proposed in this paper, which combines a pyramid vision transformer (PVT) and multi-scale fusion. This network model follows a standard encoder-decoder segmentation architecture, where PVT is used as an encoder. PVT, designed as a pure Transformer backbone for pixel-level dense prediction tasks, can consistently generate a global receptive field and, as an encoder, flexibly learn multi-scale features of medical images. Two additional modules, namely Enhanced Attention Fusion (EAF) and Edge-Enhanced Segmentation (EES) are introduced. The EAF module fuses up-sampled and skip-connected features using an attention mechanism that enhances the perception of channel and positional information. The EES module enhances the boundary features of the network through the aggregation of multi-level features of the encoder and a dynamic boundary detection operator used to obtain a boundary mask and embed it into the decoder. Extensive experiments on five datasets show that CascadeMedSeg exhibits improved performance over several state-of-the-art methods. The MIoU values for the Kvasir-SEG, CVC-ClinicDB, ISIC 2018, and BUSI datasets are 88.16, 89.79, 86.32, and 66.69%, respectively.

关键词： Medical image segmentation Pyramid vision transformer Attention mechanism Multi-scale features

来源：评论

学校读者我要写书评

暂无评论

Segmentation of Medical image Based on U-Net neural Network 24

Segmentation of Medical Image Based on U-Net Neural Network

引用

9th International Conference on Biomedical Imaging signal processing

作者： Zeng, Lingquan Xidian Univ Xian Shanxi Peoples R China

ISBN: (纸本)9798400717499

Medical image segmentation is a critical task in healthcare diagnostics and treatment planning. Among various approaches, the U-Net neural network has emerged as a revolutionary and efficient method, particularly well-suited for medical imaging applications. Its unique architecture facilitates robust feature extraction and demonstrates superior adaptability to limited datasets, which often present challenges such as indistinct anatomical boundaries. This study proposes an enhanced U-Net model that incorporates a residual module and a comprehensive skip connection mechanism. This innovative design enables the decoder to integrate multi-scale feature maps: fine-grained information from the encoder, equivalent-scale features, and broader contextual data from the decoder itself. Experimental results demonstrate that this modified U-Net architecture achieves higher segmentation accuracy and improved training efficiency compared to conventional methods.

关键词： Medical image segmentation U-net neural network residual module multi-scale feature integration image sharpening

来源：评论

学校读者我要写书评

暂无评论

Conditional Laplacian pyramid networks for exposure correction

引用

signal processing-image COMMUNICATION 2025年 134卷

作者： Huang, Mengyuan Chang, Kan Qin, Qingpao Tang, Yahui Li, Guiqing Guangxi Univ Sch Comp Elect & Informat Nanning 530004 Guangxi Peoples R China South China Univ Technol Sch Comp Sci & Engn Guangzhou 510006 Guangdong Peoples R China

Improper exposures greatly degenerate the visual quality of images. Correcting various exposure errors in a unified framework is challenging as it requires simultaneously handling global attributes and local details under different exposure conditions. In this paper, we propose a conditional Laplacian pyramid network (CLPN) for correcting different exposure errors in the same framework. It applies Laplacian pyramid to decompose an improperly exposed image into a low-frequency (LF) component and several high-frequency (HF) components, and then enhances the decomposed components in a coarse-to-fine manner. To consistently correct a wide range of exposure errors, a conditional feature extractor is designed to extract the conditional feature from the given image. Afterwards, the conditional feature is used to guide the refinement of LF features, so that a precisely correction for illumination, contrast and color tone can be obtained. As different frequency components exhibit pixel-wise correlations, the frequency components in lower pyramid layers are used to support the reconstruction of the HF components in higher layers. By doing so, fine details can be effectively restored, while noises can be well suppressed. Extensive experiments show that our method is more effective than state-of-the-art methods on correcting various exposure conditions ranging from severe underexposure to intense overexposure.

关键词： Exposure correction Low-light image enhancement Laplacian pyramid Convolutional neural network

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：