检索结果-内蒙古大学图书馆

A three-branch deep neural network for diagnosing respiratory sounds

Neural Computing and Applications 2024年第35期36卷 22611-22631页

作者： Imani, Maryam Ghassemian, Hassan Image Processing and Information Analysis Lab Faculty of Electrical and Computer Engineering Tarbiat Modares University Tehran Iran

Finding an accurate model is essential for classification of respiratory pathologies through extraction and fusion of respiratory sounds’ features. To handle the unlabeled data, a sequence of autoencoders are used for data augmentation. A deep neural network framework is proposed to extract three types of features: 1-Features based on the human auditory system, using Mel-frequency cepstral coefficients (MFCCs), 2-Temporal features contained in the sequence of sound signal using the long short-term memory (LSTM) network, and 3-The nonlinear and complex relationship among temporal characteristics in neighborhood regions using a two-dimensional (2D) convolutional neural network (CNN) applied to the sound sequence converted to a 2D time array. Three branches are fused through some fully connected layers. The ICBHI 2017 sound database is used in two cases of 6 and 3 pathological classes. In the case of 6-classes database, 98.72% overall accuracy, 98.46% kappa coefficient, 99.66% sensitivity, 98.70% specificity and 99.70% F1-Sscore is provided. In the case of 3-classes database, 97.18% overall accuracy, 95.77% kappa coefficient, 98.31% sensitivity, 99.16% specificity and 98.93% F1-score are achieved. In contrast to the many researchers that are applying CNN on the Mel spectrogram of the audios, CNN is applied to the 2D form of the time series to find the hidden relationship among the time instants in local regions. Moreover, fusion of these hidden features with the LSTM and MFCC features leads to an accurate multiple classification. This work proposes a new deep learning framework for fusion of three types of sound features, which leads to a significant improvement in multiple respiratory infection diagnosing. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

U-Net in Histological Segmentation: Comparison of the Effect of Using Different Color Spaces and Final Activation Functions

U-Net in Histological Segmentation: Comparison of the Effect...

引用

18th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2025

作者： Körmöczi, László Nyúl, László G. Department of Image Processing and Computer Graphics University of Szeged Szeged Hungary

Deepneuralnetworksbecamewidespreadinnumerousfieldsofimageprocessing, includingsemanticsegmen tation. U-Net is a popular choice for semantic segmentation of microscopy images, e.g. histological sections. In this paper, we compare the performance of a U-Net architecture in three different color spaces: the com monly used, perceptually uniform sRGB, the perceptually uniform but device-independent CIE L*a*b*, and linear RGB color space that is uniform in terms of light intensity. Furthermore, we investigate the network’s performance on data combinations that were unseen during training. © 2025 by SCITEPRESS– Science and Technology Publications, Lda.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

RepBoTNet-CESA:An Alzheimer’s Disease computer Aided Diagnosis Method Using Structural Reparameterization BoTNet and Cubic Embedding Self Attention

引用

computers, Materials & Continua 2024年第5期79卷 2879-2905页

作者： Xiabin Zhang Zhongyi Hu Lei Xiao Hui Huang College of Computer Science and Artificial Intelligence Wenzhou UniversityWenzhou325035China Key Laboratory of Intelligence Image Processing and Analysis Wenzhou UniversityWenzhou325035China

Various deep learning models have been proposed for the accurate assisted diagnosis of early-stage Alzheimer’s disease(AD).Most studies predominantly employ Convolutional Neural Networks(CNNs),which focus solely on local features,thus encountering difficulties in handling global *** contrast to natural images,Structural Magnetic Resonance Imaging(sMRI)images exhibit a higher number of channel ***,during the Position Embedding stage ofMulti Head Self Attention(MHSA),the coded information related to the channel dimension is *** tackle these issues,we propose theRepBoTNet-CESA network,an advanced AD-aided diagnostic model that is capable of learning local and global features *** combines the advantages of CNN networks in capturing local information and Transformer networks in integrating global information,reducing computational costs while achieving excellent classification ***,it uses the Cubic Embedding Self Attention(CESA)proposed in this paper to incorporate the channel code information,enhancing the classification performance within the Transformer ***,the RepBoTNet-CESA performs well in various AD-aided diagnosis tasks,with an accuracy of 96.58%,precision of 97.26%,and recall of 96.23%in the AD/NC task;an accuracy of 92.75%,precision of 92.84%,and recall of 93.18%in the EMCI/NC task;and an accuracy of 80.97%,precision of 83.86%,and recall of 80.91%in the AD/EMCI/LMCI/NC *** demonstrates that RepBoTNet-CESA delivers outstanding outcomes in various AD-aided diagnostic ***,our study has shown that MHSA exhibits superior performance compared to conventional attention mechanisms in enhancing ResNet ***,the Deeper RepBoTNet-CESA network fails to make further progress in AD-aided diagnostic tasks.

关键词： Alzheimer CNN structural reparameterization multi head self attention computer aided diagnosis

来源：评论

学校读者我要写书评

暂无评论

GridNet:efficiently learning deep hierarchical representation for 3D point cloud understanding

引用

Frontiers of computer Science 2022年第1期16卷 1-9页

作者： Huiqun WANG Di HUANG Yunhong WANG Laboratory of Intelligent Recognition and Image Processing School of Computer Science and EngineeringBeihang UniversityBeijing100191China

In this paper,we propose a novel and effective approach,namely GridNet,to hierarchically learn deep representation of 3D point *** incorporates the ability of regular holistic description and fast data processing in a single framework,which is able to abstract powerful features progressively in an efficient ***,to capture more accurate internal geometry attributes,anchors are inferred within local neighborhoods,in contrast to the fixed or the sampled ones used in existing methods,and the learned features are thus more representative and discriminative to local point *** delivers very competitive results compared with the state of the art methods in both the object classification and segmentation tasks.

关键词： 3D point clouds deep representations

来源：评论

学校读者我要写书评

暂无评论

Improved YOLOv5 Algorithm for Small Object Detection in Drone images 12th

Improved YOLOv5 Algorithm for Small Object Detection in Dr...

引用

12th International Conference on Computational Visual Media, CVM 2024

作者： Lin, Yitong Liu, Yiguang Vision and Image Processing Lab College of Computer Science Sichuan University Chengdu610065 China

ISBN: (纸本)9789819720910

The object detection in the context of drone is a hot topic in the field of computer vision in recent years. In response to the challenge of limited image feature information and the presence of numerous small and densely packed objects in drone-captured images, this paper proposes a novel feature fusion detection model, HTH-YOLOv5, based on YOLOv5. Firstly, we enhance the detection capability of small objects by adding a detection channel from high-resolution feature maps and propose a Hybrid Transformer Head (HTH) that incorporates a hybrid Transformer module, aiming to improve the network’s focus on small objects by fusing global and local feature information. Secondly, we introduce a Convolutional Attention Feature Fusion module (CA-FF) based on CBAM. This module dynamically adjusts attention weights for the allocation of original feature maps in both channel and spatial dimensions, aiming to enhance the feature extraction capability for small objects. Finally, to better capture global and contextual information, we introduce the Hybrid Transformer module into the backbone and enhance its original feature fusion method using the CA-FF module. Experiments on the Vis-Drone 2021 dataset demonstrate that, compared to the baseline YOLOv5s model, the improved model shows an increase of 7.2% in mAP50 and 6.3% in mAP75. The model trained with an input resolution of 1540×1540 achieves an mAP50 of 57.1%, marking a 12.4% improvement over YOLOv5. The improved HTH-YOLOv5 achieves increased accuracy while maintaining a detection speed of 45 FPS, making it more suitable for small object detection in drone scenarios. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Privacy Preserving Human Fall Recognition Using Human Skeleton Data

Privacy Preserving Human Fall Recognition Using Human Skelet...

引用

2023 IEEE Conference on computer Applications, ICCA 2023

作者： Htoo, Chit Kyin Sein, Myint Myint University of Computer Studies GIS and Image Processing Lab Yangon Myanmar

ISBN: (纸本)9781665435994

Fall is a critical danger for a person because he/she can loss any recontroled and stable posture and can face to any associated injuries without any caregiver. Therefore, automatic fall recognition systems have been developed to assist the person falling in short time in place. Moreover, these human-fall recognition systems have also strongly impacted from privacy and the issue of privacy is a challenge and an important societal problem for it. The aim of this paper is to propose an automatic fall recognition from human skeleton joints captured by RGB-Depth Kinect camera supporting a privacy-preservation. The recognition assesses skeleton joints of a person and measures two posture balance (body orientation and centroid height) and two gait motions (velocity and angular acceleration) of this person based on the fall and fall-like representations generated in system's virtual environment. The system has not de-identified the personal information from original captures of a person and preserves privacy which is highly impacted in fall recognition systems. Experiments on system's collected human activities using Support Vector Machine learning classification, will support to detect falls in unseen real environments with moderately high recognition accuracy. © 2023 IEEE.

关键词： Musculoskeletal system

来源：评论

学校读者我要写书评

暂无评论

FIFAWC:a dataset with detailed annotation and rich semantics for group activity recognition

引用

Frontiers of computer Science 2024年第6期18卷 271-272页

作者： Duoxuan PEI Di HUANG Yunhong WANG State Key Laboratory of Software Development Environment School of Computer Science and EngineeringBeihang UniversityBeijing 100191China Intelligent Recognition and Image Processing Lab. School of Computer Science and EngineeringBeihang UniversityBeijing 100191China

1 *** Activity Recognition(GAR),which aims to identify activities performed collectively in videos,has gained significant attention *** conventional action recognition centered on single individuals,GAR explores the c... 详细信息

关键词： has collective gained

来源：评论

学校读者我要写书评

暂无评论

Indoor Accidents Detection for Elderly Based on Features Fusion and LSTM 21

Indoor Accidents Detection for Elderly Based on Features Fus...

引用

21st IEEE International Conference on computer Applications, ICCA 2024

作者： San, Win Pa Pa Zaw, Sai Maung Maung University of Computer Studies Mandalay Image and Signal Processing Lab Mandalay Myanmar University of Computer Studies Mandalay Faculty of Computer Systems and Technology Mandalay Myanmar

ISBN: (纸本)9789998601673

Nowadays, with aging of the human society, the 'ratio of family care givers and elderly' is not equivalent and cannot give enough caring to the elderly in some countries. Therefore, automatic health monitoring system for elderly draws the interests of many researchers and become as special issue. In the implementation of elderly people monitoring system, the detection of fall events plays as an important role because the fall is the common cause which is very risky and threaten the health of elderly. The aim of this research is to implement the vision-based fall detection system which is important for health monitoring of elderly by applying the computer vision and deep learning technology. Most of the traditional vision-based fall detection systems which based on the height, width of human posture, fail to detect the fall events from different camera viewpoints. In this research, to accurately detect the fall from different camera viewpoint, the features fusion approach is applied over the deep learning model. In the implementation of the system, firstly, the human silhouette image is created by using the background subtraction. After that, the human silhouette image which have been extracted from two consecutive frames are fused as a single image called Silhouette History image (SHI). Then, the SHI results are used as shape feature. On the other hand, the motion features called Dense Optical Flow (DOF) is extracted over the two consecutive video frames and fuse SHI and DOF as a single input image. Then, the results of features fusion are feed as input for pretrained Convolution Neural Network (CNN) for extracting the deep CNN features. Finally, the Recurrent Neural Network (RNN) called Long-Short-Term-Memory (LSTM) is trained over those CNN features for recognizing the sequence of fall event. The experiments are performed using the publicly available dataset of UP-fall detection dataset for evaluating and confirming the effectiveness of the proposed method. © 2024

关键词： Fall detection

来源：评论

学校读者我要写书评

暂无评论

Training Deep Neural Classifiers with Soft Diamond Regularizers 23

Training Deep Neural Classifiers with Soft Diamond Regulariz...

引用

23rd IEEE International Conference on Machine Learning and Applications, ICMLA 2024

作者： Adigun, Olaoluwa Kosko, Bart Signal and Image Processing Institute Department of Electrical and Computer Engineering Los AngelesCA90089-2564 United States

ISBN: (纸本)9798350374889

We introduce new soft diamond regularizers that both improve synaptic sparsity and maintain classification accuracy in deep neural networks. These parametrized regularizers outperform the state-of-the-art hard-diamond Laplacian regularizer of Lasso regression and classification. They use thick-tailed symmetric alpha-stable (Sa S) bell-curve synaptic weight priors that are not Gaussian and so have thicker tails. The geometry of the diamond-shaped constraint set varies from a circle to a star depending on the tail thickness and dispersion of the prior probability density function. Training directly with these priors is computationally intensive because almost all Sa S probability densities lack a closed form. A precomputed lookup table removed this computational bottleneck. We tested the new soft diamond regularizers with deep neural classifiers on the three datasets CIFAR-10, CIFAR-100, and Caltech-256. The regularizers improved the accuracy of the classifiers. The improvements included 4.57% on CIFAR-10, 4.27% on CIFAR-100, and 6.69% on Caltech-256. They also outperformed L2 regularizers on all the test cases. Soft diamond regularizers also outperformed L1 lasso or Laplace regularizers because they better increased sparsity while improving classification accuracy. Soft-diamond priors substantially improved accuracy on CIFAR-10 when combined with dropout, batch, or data-augmentation regularization. © 2024 IEEE.

关键词： Probability density function

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Hyperspectral image Classification: Spatial and Spectral Feature Fusion with Masked Autoencoders 20

Unsupervised Hyperspectral Image Classification: Spatial and...

引用

20th CSI International Symposium on Artificial Intelligence and Signal processing, AISP 2024

作者： Mirsharji, Seyed Ali Ghassemian, Hassan Tarbiat Modares University Image Processing and Information Analysis Lab. Faculty of Electrical and Computer Engineering Tehran Iran

ISBN: (纸本)9798350383942

In this study, we present an innovative unsupervised hyperspectral image classification method using a dual-branch architecture that merges spatial and spectral feature extraction. Our unique approach employs masked autoencoders, significantly outperforming traditional methods with an impressive overall accuracy of 97.1%. The paper details the model's performance evaluation, offers visual insights into its classification capabilities, and compares it with existing techniques, demonstrating its effectiveness and potential for advancing remote sensing applications. © 2024 IEEE.

关键词： image classification

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：