检索结果-内蒙古大学图书馆

CNN Intelligent diagnosis method for bearing incipient faint faults based on adaptive stochastic resonance-wave peak cross correlation sliding sampling

引用

DIGITAL signal processing 2025年 156卷

作者： Liu, Peng Zhao, Shuo Kang, Ludi Yin, Yibing Qingdao Univ Technol Sch Mech & Automot Engn Qingdao 266525 Shandong Peoples R China Natl Univ Def Technol Lab Sci & Technol Integrated Logist Support Changsha 41007 Peoples R China

As a representative of deep learning networks, convolutional neural networks (CNN) have been widely used in bearing fault diagnosis with good results. However, the signal length and segmentation of the input CNN can have a significant impact on diagnostic accuracy. In addition, the signal-to-noise ratio of early bearing faults is usually very low, which makes it difficult for traditional CNNs to accurately identify and classify these faults. To solve this problem, this paper proposes an adaptive stochastic resonance wave peak cross-correlation sliding sampling method. Firstly, the adaptive stochastic resonance is used to reduce the noise of the original signal, and then the data is divided from the position of the signal wave peak, the correlation coefficient between the divided signals is calculated, and the maximum value is found to determine the size of the division window. Finally, it is converted into a 2D image by Gramian Angular Field and input into CNN for diagnostic classification. The design methodology was validated using the Case Western Reserve University bearing dataset. Subsequently, three validation strategies were established on a self-built platform, including mixed diagnosis of 10 different bearing states, variable speed diagnosis, and low sampling data diagnosis. The proposed method outperforms the conventional CNN by 10 % in the Case Western Reserve University dataset test set. The variable speed test set is 24.67 % and 31.17 % higher, respectively. It is 30 % higher in low sampling data diagnosis.

关键词： Early bearing faults Fault diagnosis Adaptive stochastic resonance Intecorrelation CNN

来源：评论

学校读者我要写书评

暂无评论

Medical image synthesis algorithm based on vision graph neural network with manifold matching

引用

BIOMEDICAL signal processing AND CONTROL 2025年 103卷

作者： Zeng, Xianhua Lu, Bowen Zhang, Jian Chongqing Univ Posts & Telecommun Sch Comp Sci & Technol Sch Artificial Intelligence Chongqing 400065 Peoples R China

Medical image synthesis transforms images between modalities by mapping a source image modality to a target modality. This technique provides doctors with valuable support for making diagnoses when images are unavailable and has many applications. A medical image synthesis model based on vision graph neural network with manifold matching is proposed to maintain the consistency of the image's internal and external manifold structure and improve the synthesized image's quality. When constructing a medical image synthesis model, it is important to consider the image's internal and external manifold structure. For internal manifold preservation, we construct an image graph by treating the input image's region blocks as nodes. We capture the image's internal spatial relationships by aggregating and updating the node features through a graph neural network, using the relationship information between the nodes. For external manifold preservation, a manifold corrector is introduced to correct the manifold values of the synthesized image. The manifold matching loss is then used to reduce the difference in manifold values between the synthesized and real images. The experimental results demonstrate that GMM-GAN can produce promising results as compared with other GAN-based medical image synthesis methods. Our code is available at https://***/bowenlu10/GMMGAN.

关键词： Medical image synthesis Vision graph neural network Manifold matching Manifold preservation

来源：评论

学校读者我要写书评

暂无评论

Low-dose computed tomography denoising via hybrid transformer and residual dense network

引用

JOURNAL OF INSTRUMENTATION 2025年第3期20卷 C03013-C03013页

作者： Xu, Wenting Kim, Woosung Jeon, Duhee Cho, Hyosung Yonsei Univ Dept Radiat Convergence Engn Wonju 26493 South Korea

Low-dose computed tomography (CT) has become an essential diagnostic tool, but it suffers from lower image quality and higher noise compared to normal-dose CT, leading to reduced diagnostic accuracy. To address this issue, we designed a hybrid network that leverages the capabilities of stochastic window transformer and residual dense network (RDN) for enhanced image denoising. The architecture of the hybrid network is a U-shaped network composed of encoding and decoding parts. The encoding part employs a stochastic window transformer to capture global features using a stochastic window strategy with gaussian shift, effectively reducing noise and preserving image details, while the decoding part utilizes an RDN to enhance image details by integrating information across residual structures and dense blocks, which facilitates feature reuse and improves gradient flow. Additionally, a dual-path feature enhancement module was incorporated into the proposed hybrid network to facilitate direct feature transfer between the encoding and decoding parts, ensuring to get a fusion feature by combining low- and high-level features from both paths. The experimental results demonstrated that our hybrid network significantly outperformed other existing denoising methods, achieving a peak signal-to-noise ratio of approximately 33.8, a structural similarity index measure of 0.92, and a root-mean-squared error of 8.354. Consequently, our method enhances diagnostic accuracy while reducing patient's radiation doses, which provides an effective solution for low-dose CT image denoising.

关键词： Medical-image reconstruction methods and algorithms computer-aided diagnosis image filtering image processing

来源：评论

学校读者我要写书评

暂无评论

Cascaded Thinning in Upscale and Downscale Representation for EEG signal processing

引用

IEEE TRANSACTIONS ON neural SYSTEMS AND REHABILITATION ENGINEERING 2024年 32卷 3677-3688页

作者： Doan, Quang Manh Dinh, Tran Hiep Singh, Avinash Kumar Lin, Chin-Teng Trung, Nguyen Linh VNU Univ Engn & Technol Adv Inst Engn & Technol AVITECH Hanoi 10000 Vietnam VNU Univ Engn & Technol Fac Engn Mech & Automat Hanoi 10000 Vietnam Univ Technol Sydney Australian Artificial Intelligence Inst Fac Engn & Informat Technol Sch Comp Sci Sydney NSW 2007 Australia

Smoothing filters are widely used in EEG signal processing for noise removal while preserving signals' features. Inspired by our recent work on Upscale and Downscale Representation (UDR), this paper proposes a cascade arrangement of some effective image-processing techniques for signal filtering in the image domain. The UDR concept is to visualize EEG signals at an appropriate line width and convert it to a binary image. The smoothing process is then conducted by skeletonizing the signal object to a unit width and projecting it back to the time domain. Two successive UDRs could result in a better-smoothing performance, but their binary image conversion should be restricted. The process is computationally ineffective, especially at higher line width values. Cascaded Thinning UDR (CTUDR) is proposed, exploiting morphological operations to perform a two-stage upscale and downscale within one binary image representation. CTUDR is verified on a signal smoothing and classification task and compared with conventional techniques, such as the Moving Average, the Binomial, the Median, and the Savitzky Golay filters. Simulated EEG data with added white Gaussian noise is employed in the former, while cognitive conflict data obtained from a 3D object selection task is utilized in the latter. CTUDR outperforms its counterparts, scoring the best fitting error and correlation coefficient in signal smoothing while achieving the highest gain in Accuracy (0.7640%) and F-measure (0.7607%) when used as a smoothing filter for training data of EEGNet.

关键词： Filters Electroencephalography Smoothing methods signal processing Noise Time-domain analysis Skeleton Smoothing thinning skeletonization electroen-cephalogram (EEG) signal processing time-series noise filter

来源：评论

学校读者我要写书评

暂无评论

Fall detection based on OpenPose and MobileNetV2 network

引用

IET image processing 2023年第3期17卷 722-732页

作者： Gao, Mengqi Li, Jiangjiao Zhou, Dazheng Zhi, Yumin Zhang, Mingliang Li, Bin Qilu Univ Technol Shandong Acad Sci Sch Math & Stat Jinan Peoples R China

The proposed fall detection approach is aimed at building a support system for the elders. In this work, a method based on human pose estimation and lightweight neural network is used to detect falls. First, the OpenPose is used to extract human keypoints and label them in the images. After that, the modified MobileNetV2 network is used to detect falls by integrating both human keypoint information and pose information in the original images. The above operation can use the original image information to correct the deviation in the keypoint labeling process. Through experiments, the accuracy of the proposed method is 98.6% and 99.75% on the UR and Le2i datasets, which is higher than the listed comparison methods.

关键词： Biology and medical computing feature extraction video signal processing fall detection approach original images human keypoints neural nets modified MobileNetV2 network OpenPose geriatrics Optical, image and video signal processing Computer vision and image processing techniques human keypoint information object detection image recognition listed comparison methods fall detection pose estimation Video signal processing lightweight neural network elders keypoint labeling process original image information

来源：评论

学校读者我要写书评

暂无评论

Accelerated Distributed stochastic Nonconvex Optimization Over Time-Varying Directed Networks

引用

IEEE TRANSACTIONS ON AUTOMATIC CONTROL 2025年第4期70卷 2196-2211页

作者： Chen, Yiyue Hashemi, Abolfazl Vikalo, Haris Univ Texas Austin Dept Elect & Comp Engn Austin TX 78712 USA Purdue Univ Sch Elect & Comp Engn W Lafayette IN 47907 USA

Distributed stochastic nonconvex optimization problems have recently received attention due to the growing interest of signal processing, computer vision, and natural language processing communities in applications deployed over distributed learning systems (e.g., federated learning). We study the setting where the data is distributed across the nodes of a time-varying directed network, a topology suitable for modeling dynamic networks experiencing communication delays and straggler effects. The network nodes, which can access only their local objectives and query a stochastic first-order oracle to obtain gradient estimates, collaborate to minimize a global objective function by exchanging messages with their neighbors. We propose an algorithm, novel to this setting, that leverages stochastic gradient descent with momentum and gradient tracking to solve distributed nonconvex optimization problems over time-varying networks. To analyze the algorithm, we tackle the challenges that arise when analyzing dynamic network systems that communicate gradient acceleration components. We prove that the algorithm's oracle complexity is O(1/epsilon(1.5)), and that under Polyak-Lojasiewicz condition the algorithm converges linearly to a steady error state. The proposed scheme is tested on several learning tasks: a nonconvex logistic regression experiment on the MNIST dataset, an image classification task on the CIFAR-10 dataset, and an NLP classification test on the IMDB dataset. We further present numerical simulations with an objective that satisfies the PL condition. The results demonstrate superior performance of the proposed framework compared to the existing related methods.

关键词： stochastic processes Optimization signal processing algorithms Convergence Computational modeling Heuristic algorithms Complexity theory Linear programming Distance learning Directed graphs Decentralized nonconvex optimization stochastic nonconvex optimization time-varying directed network

来源：评论

学校读者我要写书评

暂无评论

Maximum-correntropy-based sequential method for fast neural population activity reconstruction in the cortex from incomplete abnormally-disturbed noisy measurements

引用

COMMUNICATIONS IN NONLINEAR SCIENCE AND NUMERICAL SIMULATION 2024年 139卷

作者： Kulikova, M. V. Kulikov, G. Yu. Lisbon Univ Inst Super Tecn CEMAT Ctr Computat & Stochast Math Ave Rovisco Pais 1 P-1048001 Lisbon Portugal

This paper continues to explore the membrane potential reconstruction and pattern recognition problem in a neural tissue modeled by stochastic Dynamic neural Field (SDNF) equation. Although recent research has suggested an efficient solution based on the state-space approach through nonlinear Bayesian filtering framework, it is becoming extremely difficult to ignore the existence of non-Gaussian uncertainties in the SDNFs as well as the stability problem of neuronal population dynamics to outliers. Motivated by recent events in signal processing and mathematical neuroscience, this paper explores the SDNFs in a presence of non-Gaussian uncertainties, which is the shot noise case, where the corrupted data might appear due to broken sensors. We derive the "distributionally robust'' state estimator for the membrane potential reconstruction process that is the Maximum Correntropy Criterion Extended Kalman Filter (MCC-EKF) as well as its fast and numerically robust (to roundoff) implementation method by using the sequential principle of processing the measurement vectors. The numerical experiments are provided to illustrate the performance of the novel estimation methods.

关键词： stochastic dynamic neural fields Maximum correntropy criterion Non-gaussian uncertainties Extended kalman filter Sequential filtering

来源：评论

学校读者我要写书评

暂无评论

Stable successive neural image Compression via coherent demodulation-based transformation

引用

signal processing 2025年 227卷

作者： Bao, Youneng Tan, Wen Li, Mu Meng, Fanyang Liang, Yongsheng Harbin Inst Technol Shenzhen Sch Elect & Informat Engn Pingshan Rd Shenzhen 518055 Guangdong Peoples R China Harbin Inst Technol Shenzhen Sch Comp Sci Pingshan Rd Shenzhen 518055 Guangdong Peoples R China Shenzhen Technol Univ Coll Big Data & Internet Shenzhen 518118 Guangdong Peoples R China Peng Cheng Lab Shenzhen 518000 Guangdong Peoples R China

neural image Compression (NIC) has made significant strides in recent years. However, the existing NIC methods demonstrate instability issues during iterative re-compression cycles, which can degrade image quality with each cycle. This paper introduces a novel framework aimed at enhancing the stability of NIC methods. We first conducted a theoretical analysis and identified that the instability in current NIC methods stems from a lack of idempotency in transformations. Drawing from the domain of signal processing, we then examined the principles of idempotency incoherent demodulation techniques. This examination led to the identification of three foundational principles that inform the design of stable transformations: the cosine function, parameter sharing, and low-pass filtering. Leveraging these insights, we propose the innovative Coherent Demodulation- based Transformation (CDT), which is designed to address the stability challenges in NIC by incorporating these principles into its architecture. The experimental results suggest that CDT not only significantly improve the re-compression stability but also preserves the codec's rate-distortion performance. Furthermore, it can be broadly applied in current NIC structures. The effectiveness of the module endorses the viability of designing transformation networks based on Coherent Demodulation principles, playing a crucial role in enhancing stability of NIC. The code will be available at https://***/baoyu2020/Stable_SuccessiveNIC.

关键词： Deep learning Communication system image compression Idempotent

来源：评论

学校读者我要写书评

暂无评论

Resnet 152 convolutional adaptive spiral flying sparrow search neural network for image enhancement of video surveillance scene

引用

signal image AND VIDEO processing 2025年第3期19卷 1-11页

作者： Chellam, J. Angel Ida Malliga, P. Manoharan, Mathankumar Ramkumar, M. Sri Ramakrishna Engn Coll Dept Informat Technol Coimbatore 641022 Tamil Nadu India Natl Inst Tech Teachers Training & Res Dept Comp Sci & Engn Chennai 600113 India SRM Inst Sci & Technol Sch Elect & Elect Engn Tiruchirappalli India Sri Krishna Coll Engn & Technol Dept Elect & Commun Engn Coimbatore 641008 Tamil Nadu India

In the field of video surveillance, effective image enhancement is pivotal for extracting valuable information from challenging visual environments. Enhancing images from video surveillance scenes is challenging due to varying lighting conditions ranging from bright daylight to low-light or nighttime settings. Noise, artifacts and distortions in video frames further degrade quality, while real-time processing requirements add complexity. To overcome these issues, this research focuses on developing a specialized neural network tailored for enhancing images captured in video surveillance scenarios. The primary objective is to significantly boost the visual quality of surveillance video frames. To achieve both accuracy and efficiency, Convolutional neural Network (CNN) based on ResNet-152, is specifically designed for enhancing images in video surveillance settings. The research aims to enhance adaptability to varying lighting conditions, weather patterns and scene complexities. Uniform frame sampling (UFS) ensures simplicity in implementation and computational efficiency by consistently extracting frames at regular intervals. To further enhance the performance of the ResNet-152 CNN, an Adaptive Spiral Flying Sparrow Search Algorithm (ASFSSA) is employed. Experimental outcomes reveal that the proposed system outperforms traditional approaches, achieving impressive metrics like accuracy of 98%, recall of 95%, precision of 95.7%, F1-score of 98%, specificity of 96%, sensitivity of 97.8% and a peak signal-to-noise ratio of 35%. Additionally, the structural similarity index measure, root mean square error and mean average error of the proposed technique are reported at 0.09%, 4% and 3.02% respectively, showcasing improvements over current methods.

关键词： Uniform frame sampling Adaptive spiral flying sparrow search algorithm Convolutional neural network ResNet-152 Video surveillance

来源：评论

学校读者我要写书评

暂无评论

Robust image Fingerprinting Algorithm Based on Local and Global Content Dependencies

引用

IEEE signal processing LETTERS 2024年 31卷 336-340页

作者： Li, Yuenan Zhang, Wei Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China

image fingerprinting summarizes the unique visual characteristics of an image into a robust and compact ID for content identification. This technique is widely adopted by social networks to identify unauthorized uploads of copyrighted content. In this letter, we propose a deep neural network based image fingerprinting algorithm, where a neural network is designed to capture the short and long-range structural dependencies of an image and compress the representative features into fingerprints. The training algorithm optimizes the content identification accuracy of the fingerprinting model from a hypothesis-testing perspective. We propose a differentiable training objective for minimizing the error rate of the hypothesis-testing problem. Since real applications prefer binary fingerprints, we also develop an adversarial training scheme to progressively force the outputs of the neural network to approach binary states, aiming to minimize the performance loss caused by fingerprint binarization. The experimental results show that the proposed algorithm achieves more accurate content identification than state-of-the-art methods and is insensitive to fingerprint binarization.

关键词： image fingerprinting perceptual hashing content identification 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：