检索结果-内蒙古大学图书馆

11th Frontier Academic Forum of Electrical Engineering, FAFEE 2024

作者： Xinguang, Liu Baoling, Liu Jun, He Xixi, Liu Yulong, Yuan Xiaocui, Yuan Yongtao, Wang Nanchang Institute of Technology Jiangxi Nanchang330099 China State Grid Jiangxi Electric Power Research Institute Jiangxi Nanchang330096 China

ISBN: (纸本)9789819788194

The implementation of deep learning-based fault diagnosis methodologies has been increasingly observed across diverse sectors within the power industry. This is particularly relevant in contexts where power stations generate vast quantities of operational data that necessitate advanced real-time processing capabilities. However, it has been observed that the efficacy of isolated deep learning models often falls below expectations, primarily due to their limited generalization capabilities. Leading to their limited application in the fault detection of power station automatic control loops. In light of these challenges, the present study introduces an innovative anomaly detection methodology specifically designed for control loops. Based on deep clustering and transfer learning. Initially, the methodology employs an autoencoder clustering algorithm to systematically categorize the operational conditions prevalent in control loops. Subsequently, the VAE-LSTM (Variational Auto Encoder-Long Short Term Memory) model is deployed to meticulously extract the latent features of the difference sequences between controlled parameters and set values. The source domain model undergoes training through the minimization of the loss function, thereby optimizing its parameters for enhanced performance. Finally, by harnessing the principles of transfer learning, the model undergoes fine-tuning of its network parameters by training the feature distribution distance of the LSTM network with target domain data, This process significantly enhances the efficiency and accuracy of fault diagnosis. Experimental validation has confirmed the method's superior performance across datasets from various domains. Moreover, it has demonstrated the capability to conduct real-time fault diagnosis across a multitude of automatic control loops. © Beijing Paike Culture Commu. Co., Ltd. 2025.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

LIGHTWEIGHT DUAL-DOMAIN NETWORK FOR real-time MEDICAL image SEGMENTATION 29

LIGHTWEIGHT DUAL-DOMAIN NETWORK FOR REAL-TIME MEDICAL IMAGE ...

引用

IEEE International Conference on image processing (ICIP)

作者： Liu, Libo Fan, Xinxin Zhang, Xiaodong Hu, Qingmao Chinese Acad Sci Shenzhen Inst Adv Technol Shenzhen Peoples R China Univ Chinese Acad Sci Sch Artificial Intelligence Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China

ISBN: (数字)9781665496209

ISBN: (纸本)9781665496209

With the development of deep learning, deep convolution neural networks for medical image segmentation tasks have become more and more complex in pursuit of higher accuracy. In most scenarios, medical image segmentation pursues accuracy rather than speed, However, real-time performance is crucial in some scenarios, such as surgical navigation and diagnosis of acute stroke. So design of high-precision, lightweight and real-time medical image segmentation network has become an urgent need. To this end, a novel lightweight dual-domain network (LDD-Net) has been proposed in this paper. LDD-Net is comprised of two branches, learning respectively from the frequency domain and the spatial domain. In the frequency domain branch, the image spatial resolution is compressed via discrete cosine transform to have a large receptive field, so that better semantic context features can be learned. In the spatial domain branch, high-resolution feature representations with more details are learned. Finally, the learned features of these two branches are fused to yield high accuracy with low computational cost. The proposed method has been validated on two medical image segmentation datasets to yield the state-of-the-art performances with greatly reduced inference time and parameters of the learned models.

关键词： Medical image segmentation deep learning lightweight convolution neural networks lightweight dual-domain network

来源：评论

学校读者我要写书评

暂无评论

RHNet: A real-time High-Resolution Network for Surface Defect Detection 43

RHNet: A Real-Time High-Resolution Network for Surface Defec...

引用

43rd Chinese Control Conference, CCC 2024

作者： Wang, Yijie Ren, Zhigang Cai, Jianpu Wu, Zongze Xie, Shengli Shenzhen518060 China Guangdong University of Technology School of Automation Key Laboratory of Intelligent Information Processing and System Integration of IoT Ministry of Education Guangzhou510006 China Guangdong University of Technology Guangdong-HongKong-Macao Joint Laboratory for Smart Discrete Manufacturing Guangzhou510006 China Guangzhou510006 China

ISBN: (纸本)9789887581581

In the industrial domain, surface defect detection after multiple processing steps is crucial for improving the outgoing quality of products. However, due to the characteristics of surface defects, such as low contrast, irregular shapes, small and elongated features, and significant noise. To address these issues, this paper proposes a novel real-time high-resolution network (RHNet). Unlike previous models, RHNet emphasizes the importance of modeling spatial information to capture small defects. Therefore, the output feature maps of each module are maintained at 1/4 resolution of the original image. We introduce a short-term dual-branch module (SDBM) to handle real-time processing of high-resolution feature maps. Additionally, FPAPPM is proposed for swift extraction of deep information and multi-scale context. Through extensive experiments, we demonstrate that our proposed model exhibits robust defect modeling capabilities and detection performance while still meeting real-time requirements. © 2024 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

An Improved Seq-deepfake Detection Method 1st

An Improved Seq-Deepfake Detection Method

引用

1st International Conference on Applied Intelligence, ICAI 2023

作者： Deng, Zhenrong You, Kang Yang, Rui Hu, Xinru Chen, Yuren Guangxi Key Laboratory of Images and Graphics Intelligent Processing Guilin University of Electronic Technology Guilin541004 China Guangxi Construction Industry Mostly Guangxi China

ISBN: (纸本)9789819709021

Facial processing technology's ability to generate realistic human faces poses significant societal risks when exploited maliciously. deep face fraud detection relies on deep learning to meticulously scrutinize the manipulation sequence of fake faces, uncovering deceptive traces. This study focuses on Detecting Sequential deepFake Operations (Seq-deepFake), transforming the face detection task into an image-to-sequence *** enhance detection accuracy, this paper introduces a Seq-deepFake detection method. The Seq-deepFake Transformer model's activation function is refined, incorporating the Rectified Randomized Leaky Unit (RReLU) to address learning rate challenges associated with negative input values. Furthermore, diverse attention mechanism modules are integrated into the backbone network, forming the innovative CLSP-Resnet-50 model. Experimental results demonstrate the efficacy of the enhanced Seq-deepFake model, employing two evaluation metrics on a deepfake dataset, showcasing improved accuracy. Comparative analysis against other real and fake face detection methods substantiates the effectiveness of the Seq-deepFake model. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Fake detection

来源：评论

学校读者我要写书评

暂无评论

real-time denoising enables high-sensitivity fluorescence time-lapse imaging beyond the shot-noise limit

引用

NATURE BIOTECHNOLOGY 2023年第2期41卷 282-+页

作者： Li, Xinyang Li, Yixin Zhou, Yiliang Wu, Jiamin Zhao, Zhifeng Fan, Jiaqi Deng, Fei Wu, Zhaofa Xiao, Guihua He, Jing Zhang, Yuanlong Zhang, Guoxun Hu, Xiaowan Chen, Xingye Zhang, Yi Qiao, Hui Xie, Hao Li, Yulong Wang, Haoqian Fang, Lu Dai, Qionghai Tsinghua Univ Dept Automat Beijing Peoples R China Tsinghua Univ Tsinghua Shenzhen Int Grad Sch Shenzhen Peoples R China Tsinghua Univ Inst Brain & Cognit Sci Beijing Peoples R China Hangzhou Zhuoxi Inst Brain & Intelligence Hangzhou Peoples R China Fudan Univ Sch Informat & Technol Shanghai Peoples R China Tsinghua Univ Beijing Key Lab Multidimens & Multiscale Computat Beijing Peoples R China Tsinghua Univ IDG McGovern Inst Brain Res Beijing Peoples R China Tsinghua Univ Dept Elect Engn Beijing Peoples R China Peking Univ Sch Life Sci State Key Lab Membrane Biol Beijing Peoples R China PKU IDG McGovern Inst Brain Res Beijing Peoples R China

deepCAD-RT denoises fluorescence time-lapse images in real time. A fundamental challenge in fluorescence microscopy is the photon shot noise arising from the inevitable stochasticity of photon detection. Noise increases measurement uncertainty and limits imaging resolution, speed and sensitivity. To achieve high-sensitivity fluorescence imaging beyond the shot-noise limit, we present deepCAD-RT, a self-supervised deep learning method for real-time noise suppression. Based on our previous framework deepCAD, we reduced the number of network parameters by 94%, memory consumption by 27-fold and processing time by a factor of 20, allowing real-time processing on a two-photon microscope. A high imaging signal-to-noise ratio can be acquired with tenfold fewer photons than in standard imaging approaches. We demonstrate the utility of deepCAD-RT in a series of photon-limited experiments, including in vivo calcium imaging of mice, zebrafish larva and fruit flies, recording of three-dimensional (3D) migration of neutrophils after acute brain injury and imaging of 3D dynamics of cortical ATP release. deepCAD-RT will facilitate the morphological and functional interrogation of biological dynamics with a minimal photon budget.

关键词： Fluorescence imaging image processing Microscopy Software

来源：评论

学校读者我要写书评

暂无评论

deep learning based Dysgraphia Symptom Prediction 15

Deep Learning based Dysgraphia Symptom Prediction

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Vanjari, Nisha Shete, Prasanna K. J. Somaiya Institute of Technology Mumbai India K. J. Somaiya College of Engineering Mumbai India

ISBN: (纸本)9798331300579

People who have learning disabilities frequently struggle in the areas of reading and writing. The main effects of learning disabilities are bad grades and a lack of motivation that lasts a lifetime. Most of the time, it starts in childhood and has a big effect on academics. The early recognition by using technological prediction will assist the academicians to devise special learning strategy to accommodate such children in the teaching learning and hence perform well in academics. Specific learning disability is a complex neurodevelopmental/neurological disorder that affects many children. learning Disability have two major categories like Specific learning Disability in reading (dyslexia) and Specific learning Disability in writing (dysgraphia). Our primary focus is to predict the symptoms of dysgraphia using deep learning algorithms. deep learning algorithms are general-purpose methods of artificial intelligence that can learn patterns from data without the need to define them apriori. Convolutional Neural Network is a type of deep learning algorithm that is particularly well-suited for image recognition and processing tasks. To generate automated and trained models that reduce the amount of human intervention by significant margins is the primary goal of deep learning. For experimentation purpose, we have used MNIST alphabets dataset to analyse the problem. For this research, we have reviewed several methods like handwriting analysis, eye tracking, Tests, questionnaire, games;etc but we choose handwriting modality for our research. In this paper, experimental study done on convolutional neural network and transfer learning models for predict the dysgraphia symptoms and implemented using publicly available dataset of MNIST alphabet dataset. We compared the accuracy of the VGG16, VGG19, and ResNet50 architectures. Based on the analysis, we have arrived at the conclusion that the VGG16 architecture is the most superior option in training and validation phase. © Gr

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Performance Optimization and Model Research of Machine learning in Speech Recognition System 6

Performance Optimization and Model Research of Machine Learn...

引用

6th International Conference on Artificial Intelligence and Computer Applications, ICAICA 2024

作者： Shang, Wenwen Xi'an FANYI University Shaanxi Xi'an710105 China

ISBN: (纸本)9798350377255

This paper proposes an improved method based on machine learning, which combines the deep neural network (DNN) architecture and speech enhancement technology to significantly improve the recognition accuracy and robustness of the system. In traditional speech recognition models, complex noise environments often lead to inaccurate feature extraction and poor model training effects, thereby reducing the practical application value of the system. To solve the above problems, this paper introduces adaptive noise reduction algorithms and time-frequency domain enhancement technologies in speech preprocessing to optimize the quality of input signals. In addition, through the improved DNN model structure, the ability to model the spatiotemporal features of speech signals is enhanced. At the same time, the dynamic learning rate strategy and gradient optimization algorithm are adopted to improve the convergence efficiency and generalization performance of the model. Experiments verify the superior performance of the improved model in multi-noise scenarios. Compared with traditional methods, the speech recognition accuracy of the improved model is improved by 12.5%, indicating that it has stronger robustness under noise interference. The processing delay of the system is reduced by an average of 15.3%, which greatly improves the real-time performance. © 2024 IEEE.

关键词： Frequency domain analysis

来源：评论

学校读者我要写书评

暂无评论

LeNet - IoT Security Aware Intrusion Detection Framework Using deep learning

LeNet - IoT Security Aware Intrusion Detection Framework Usi...

引用

2024 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems, ICSES 2024

作者： Ananthajothi, K. Adharshini, C. Amirdha, M. Rajalakshmi Engineering College Department of Computer Science and Engineering Chennai India

ISBN: (数字)9798331543624

ISBN: (纸本)9798331543617

The rising complexity of cyberthreats created the demand for compelling cybersecurity solutions that along with expanding Internet of Things (IoT) devices has made them indispensable. In the work our proposal is a deep learning-Assisted Intrusion Detection System (IDS), based on the LeNet architecture, that can be used for the identification and prevention of potential intrusions in Internet of Things environments. To increase the detection accuracy, we developed our method based on a properly structured preprocessing step comprising normalisation and data augmentation using image-based and QR-based datasets. We take a look at the architecture of LeNet including the pooling and convolutional layers that help to identify important features of the input dataset. The final parts of the model are fully linked layers and a Softmax classifier that predicts the type of anomaly. Performance metrics measure the efficacy of the IDS including precision, recall, accuracy, and F1 score that reveal our method surpass the classical methods with a high margin. We shall describe in this paper how one can train the LeNet model using IoT-specific attack data so that it could be used to detect intrusion in surrounding network space and inject the walled-off attack, thereby facilitating threat prevention. Since the model is GPU accelerated, it converged close to what was able to protect the IoT network effectively from the evolving cyber-attacks and was quite close to its high accuracy and real time efficient performance in terms of time. © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Improved Wild Horse Optimizer with deep learning Model for Skin Lesion Detection and Classification on Dermoscopic images 23th

Improved Wild Horse Optimizer with Deep Learning Model for S...

引用

23rd International Conference on Intelligent Systems Design and Applications, ISDA 2023

作者： Maranco, M. Tyagi, Amit Kumar Sivakumar, M. Department of Networking and Communications School of Computing Faculty of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chengalpattu 603203 India Department of Fashion Technology National Institute of Fashion Technology New Delhi India

ISBN: (纸本)9783031648120

Melanoma is a deadly kind of skin cancer which can spread to other parts of the body. Therefore, it is necessary to identify melanoma at the beginning level. Visual examinationat the time of medical examination of skin lesion is a challenging process as there exist high resemblance among the lesions. Besides, dermoscopy is a non-invasive imaging tool which enables visualizationof the skin surface by light magnifying device and immersion fluid. Traditional image processing models such as histogram thresholding, clustering, or active contours are employed to segment skin lesions. This motivates the study of how to best employ AI, and deep learning (DL) algorithms in particular, to identify skin lesions in dermoscopic pictures. By using these images, current research refines the existing wild horse optimizer (IWHO) by using a deep learning driven skin lesion detection and classification (IWHODL-SLDC) *** proposed model utilizes optimal Tsallis entropy based image segmentation. Besides, Squeeze Netapproach can be employed to derive feature vectors. Also, IWHO with deep wavelet neural network (DWNN) technique is used to classify the dermoscopic images. Numerous tests are carried out to guarantee the enhancement of the IWHODL-SLDC method on the reference dataset. It was found via experimentation that the IWHODL-SLDC model outperformed the other approaches in a wide range of ways. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： image classification

来源：评论

学校读者我要写书评

暂无评论

Self-supervised pre-training improves fundus image classification for diabetic retinopathy

Self-supervised pre-training improves fundus image classific...

引用

Conference on real-time image processing and deep learning

作者： Lee, Joohyung Lee, Eung-Joo Korea Elect Technol Inst 9fl Elect Ctr 11 Worldcup Bukro 54 Gil Mapo Gu Seoul South Korea MGH Dept Radiol CAMCA Boston MA USA Harvard Med Sch Boston MA 02115 USA

ISBN: (纸本)9781510650817;9781510650800

This paper assesses the efficacy of self-supervised learning in the deepDR Diabetic Retinopathy image Dataset (deepDRiD). Recently, self-supervised learning has achieved great success in the field of Computer Vision. Particularly, self-supervised learning can effectively serve the field of medical imaging where a large amount of labeled data is usually limited. In this paper, we apply the Bootstrap Your Own Latent (BYOL) approach to grade diabetic retinopathy which scores the lowest among the MedMNIST dataset. With the pre-trained model using BYOL, we evaluate the efficacy of the BYOL approach on deepDRiD following fine-tuning protocols. Further, we compare the performance of the model with the model from scratch and proved the effectiveness of BYOL in deepDRiD. Our experiment shows that BYOL can boost the performance of grading diabetic retinopathy.

关键词： Self-supervised learning Diabetic Retinopathy MedMNIST BYOL deep learning

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：