检索结果-内蒙古大学图书馆

A hybrid TCN-BiLSTM short-term load forecasting model for ship electric propulsion systems combined with multi-step feature processing

引用

OCEAN ENGINEERING 2025年 316卷

作者： Pang, Shuo Zou, Liang Zhang, Li Wang, Hui Wang, Yawei Liu, Xingdou Jiang, Jundao Shandong Univ Sch Elect Engn Jinan 250061 Peoples R China Shanghai Jiao Tong Univ Sch Elect Informat & Elect Engn Shanghai 200240 Peoples R China

The target of load forecasting in this paper is the output power of electric propulsion motor, which can provide reference for the real-time optimization of ship energy management. In electric propulsion ships, the characteristics of high proportion of propulsion load, high dynamic and strong environmental dependence pose challenges to accurate propulsion load prediction. In this paper, a short-term prediction model of hydrogen ship propulsion load based on Temporal Convolutional Network (TCN) and Bidirectional Long Short-Term Memory (BiLSTM) is proposed. Firstly, the temporal features of long historical distance in ship navigation are extracted by TCN. Then, the temporal features are used as the input of BiLSTM, while learning the forward and backward information of the temporal features ,and mine the potential time correlation of the propulsion load. Finally, the prediction result is mapped into the sample space through fully connected layer. In data processing, Variable Modal Decomposition (VMD) is used to separate the multicomponent and nonstationary component features from the original sequence. Then, the data dimension is reduced through Principal Component Analysis (PCA) to retain the effective information characteistics. In model training, K-means clustering is used to achieve automatic data classification to improve the prediction performance. Based on navigational and hydrometeorological dataset of a hydrogen ship, this paper conducts comparative simulation experiments on multiple prediction time scales for three ship conditions: normal, working and in/out-of-port conditions. Compared with other models, the experimental results show the performance superiority of the proposed method in various operational conditions.

关键词： Short-term load forecasting Hydrogen fueled ships Deep learning feature processing

来源：评论

学校读者我要写书评

暂无评论

Texture classification-based feature processing for violence-based anomaly detection in crowded environments

引用

IMAGE AND VISION COMPUTING 2022年 124卷

作者： Mohamed, Abdallah A. Alqahtani, Fayez Shalaby, Ahmed Tolba, Amr Menoufia Univ Fac Sci Math & Comp Sci Dept Shibin Al Kawm 32511 Egypt King Saud Univ Coll Comp & Informat Sci Software Engn Dept Riyadh 12372 Saudi Arabia Univ Louisville Speed Sch Engn Bioengn Dept Louisville KY 40292 USA King Saud Univ Community Coll Comp Sci Dept Riyadh 11437 Saudi Arabia

Anomaly detection from video surveillance inputs helps to improve security in crowded places and outdoors. The captured image is analyzed to identify human faces, objects, and abnormal events through computer-aided analytics. This article proposes a Texture-Classification-based feature processing (TCFP) technique for distinguishing anomalies in captured video inputs. The anomalies are identified as events from the sequence frames wherein the dynamic inputs are distinguished using their features. Deep learning is employed for temporal training features based on frame characteristics in this distinguishing process. The input frame is segregated using textural boundaries separated using non-dimensional features. The learning process trains dimensional and non-dimensional features for identifying anomalies and maximizing detection accuracy. The textural boundaries are defined using the non-dimensional vectors present in the frame series in the different face classifications. Therefore, the errors are confined within selective boundaries without impacting the preceding feature. This improves the F1score with less processing time.(c) 2022 Elsevier B.V. All rights reserved.

关键词： Anomaly detection Deep learning feature processing Textural analysis Video surveillance

来源：评论

学校读者我要写书评

暂无评论

A short-term electricity consumption forecasting approach based on feature processing and hybrid modelling

引用

IET GENERATION TRANSMISSION & DISTRIBUTION 2022年第10期16卷 2003-2015页

作者： Wei, Minjie Wen, Mi Luo, Junran Shanghai Univ Elect Power Coll Elect Engn Shanghai Peoples R China Shanghai Univ Elect Power Coll Comp Sci & Technol 1851 Huchenghuan Rd Shanghai Peoples R China

The short-term electricity consumption forecasting can help to ensure the safe and reliable operation of the power system. Power companies usually need to report the electricity consumption of the current month five to seven days in advance and make a power generation plan for the next month. The existing studies are usually lack of appropriate feature selection methods and hard to achieve satisfactory results. This paper proposes a short-term electricity consumption forecasting approach based on feature processing and hybrid modelling. The maximum information coefficient (MIC) is employed to analyse the feature correlation, the electricity consumption curves are converted to several sub-sequences of different frequency bands by the variational mode decomposition (VMD) to describe signal characteristics accurately, a hybrid model based on bidirectional gated recurrent unit (BiGRU) is innovated to extract the temporal and spatial features of the data and capture the contextual information from the complete time series, attention mechanism is used to do extract useful information and assign weights to make forecast. Compared with several benchmark methods, the proposed approach achieves better electricity consumption curve fitting and higher forecasting accuracy with the increase of forecasting step size.

关键词： load forecasting feature selection methods VMD maximum information coefficient frequency bands contextual information curve fitting hybrid modelling power system bidirectional gated recurrent unit MIC feature selection power consumption feature correlation power generation plan variational mode decomposition safe reliable operation electricity supply industry power companies BiGRU feature extraction signal characteristics electricity consumption curve fitting short-term electricity consumption forecasting power generation planning time series power engineering computing power generation reliability feature processing

来源：评论

学校读者我要写书评

暂无评论

Alpha-Band Phase Modulates Bottom-up feature processing

引用

CEREBRAL CORTEX 2022年第6期32卷 1260-1268页

作者： Jia, Jianrong Fan, Ying Luo, Huan Hangzhou Normal Univ Ctr Cognit & Brain Disorders Affiliated Hosp Hangzhou 310015 Zhejiang Peoples R China Hangzhou Normal Univ Inst Psychol Sci 2318 Yuhangtang Rd Hangzhou 311121 Zhejiang Peoples R China Hangzhou Normal Univ Zhejiang Key Lab Res Assessment Cognit Impairment Hangzhou 311121 Zhejiang Peoples R China Peking Univ Sch Psychol & Cognit Sci 5 Yiheyuan Rd Beijing 100871 Peoples R China Peking Univ IDG McGovern Inst Brain Res Beijing 100871 Peoples R China Peking Univ Beijing Key Lab Behav & Mental Hlth Beijing 100871 Peoples R China

Recent studies reveal that attention operates in a rhythmic manner, that is, sampling each location or feature alternatively over time. However, most evidence derives from top-down tasks, and it remains elusive whether bottom-up processing also entails dynamic coordination. Here, we developed a novel feature processing paradigm and combined time-resolved behavioral measurements and electroencephalogram (EEG) recordings to address the question. Specifically, a salient color in a multicolor display serves as a noninformative cue to capture attention and presumably reset the oscillations of feature processing. We then measured the behavioral performance of a probe stimulus associated with either high- or low-salient color at varied temporal lags after the cue. First, the behavioral results (i.e., reaction time) display an alpha-band (similar to 8 Hz) profile with a consistent phase lag between high- and low-salient conditions. Second, simultaneous EEG recordings show that behavioral performance is modulated by the phase of alpha-band neural oscillation at the onset of the probes. Finally, high- and low-salient probes are associated with distinct preferred phases of alpha-band neural oscillations. Taken together, our behavioral and neural results convergingly support a central function of alpha-band rhythms in feature processing, that is, features with varied saliency levels are processed at different phases of alpha neural oscillations.

关键词： alpha-band behavioral oscillation feature processing feature-based attention phase modulation

来源：评论

学校读者我要写书评

暂无评论

EAFP-Med: An efficient adaptive feature processing module based on prompts for medical image detection

引用

EXPERT SYSTEMS WITH APPLICATIONS 2024年 247卷

作者： Li, Xiang Lan, Long Lahza, Husam Yang, Shaowu Wang, Shuihua Yang, Wenjing Liu, Hengzhu Zhang, Yudong Natl Univ Def Technol Coll Comp Sci & Technol Changsha 410073 Peoples R China Natl Univ Def Technol Inst Quantum Informat Changsha 410073 Peoples R China Natl Univ Def Technol Coll Comp Sci & Technol State Key Lab High Performance Comp Changsha 410073 Peoples R China King Abdulaziz Univ Fac Comp & Informat Technol Dept Informat Technol Jeddah 21589 Saudi Arabia Xian Jiaotong Liverpool Univ Dept Biol Sci Suzhou 215123 Jiangsu Peoples R China Univ Leicester Sch Comp & Math Sci Leicester LE1 7RH England Southeast Univ Sch Comp Sci & Engn Nanjing 210096 Jiangsu Peoples R China Univ Liverpool Dept Math Sci Liverpool L69 3BX England

The rapid proliferation of medical imaging technologies presents a significant challenge for cross-domain adaptive image detection, as lesion representations can vary dramatically across technologies. To address this issue, we draw inspiration from large language models to propose EAFP-Med, an efficient adaptive feature processing module based on prompts for medical image detection. EAFP-Med incorporates a prompt-driven dynamic parameter update mechanism, empowering it to extract cross-domain multi-scale lesion features from medical images of diverse modalities adaptively. This exceptional flexibility liberates it from the constraints of any particular imaging technique, fostering great adaptability. Furthermore, EAFP-Med can also serve as a feature preprocessing module connected to any model front-end to enhance the lesion features in input images. Moreover, we propose a novel adaptive disease detection model named EAFP-Med ST, which utilizes the Swin Transformer V2 - Tiny (SwinV2-T) as its backbone and connects it to EAFP-Med. We have compared our method to nine state-of-the-art methods. Experimental results show that the overall accuracy of EAFP Med ST on chest Xray, brain magnetic resonance imaging, and skin image datasets is 98.47 %, 97.60 %, and 99.06 %, respectively, superior to all the compared state-of-the-art methods.

关键词： Adaptive Detection Cross-Domain feature processing Medical Images Prompt

来源：评论

学校读者我要写书评

暂无评论

Convolutional Graph Neural Network with Novel Loss Strategies for Daily Temperature and Precipitation Statistical Downscaling over South China

引用

Advances in Atmospheric Sciences 2025年第1期42卷 232-247页

作者： Wenjie YAN Shengjun LIU Yulin ZOU Xinru LIU Diyao WEN Yamin HU Dangfu YANG Jiehong XIE Liang ZHAO Central South University Changsha 410083China Hunan Expressway Information Technology Co. Ltd.Changsha 410053China Guangdong Climate Center Guangzhou 510610China Hong Kong Polytechnic University Shenzhen Research Institute Guangdong 518057China Jieyang Meteorological Bureau Jieyang 522031China State Key Laboratory of Numerical Modeling for Atmosphere Sciences and Geophysical Fluid Dynamics(LASG) Institute of Atmospheric PhysicsChinese Academy of SciencesBeijing 100029China

Traditional meteorological downscaling methods face limitations due to the complex distribution of meteorological variables,which can lead to unstable forecasting results,especially in extreme *** overcome this issue,we propose a convolutional graph neural network(CGNN)model,which we enhance with multilayer feature fusion and a squeeze-and-excitation ***,we introduce a spatially balanced mean squared error(SBMSE)loss function to address the imbalanced distribution and spatial variability of meteorological *** CGNN is capable of extracting essential spatial features and aggregating them from a global perspective,thereby improving the accuracy of prediction and enhancing the model's generalization *** on the experimental results,CGNN has certain advantages in terms of bias distribution,exhibiting a smaller *** it comes to precipitation,both UNet and AE also demonstrate relatively small *** for temperature,AE and CNNdense perform outstandingly during the *** time correlation coefficients show an improvement of at least 10%at daily and monthly scales for both temperature and ***,the SBMSE loss function displays an advantage over existing loss functions in predicting the98th percentile and identifying areas where extreme events ***,the SBMSE tends to overestimate the distribution of extreme precipitation,which may be due to the theoretical assumptions about the posterior distribution of data that partially limit the effectiveness of the loss *** future work,we will further optimize the SBMSE to improve prediction accuracy.

关键词： statistical downscaling convolutional graph neural network feature processing SBMSE loss function

来源：评论

学校读者我要写书评

暂无评论

Priority-based Multi-feature Vector Model Using Convolution Neural Network for Biometric Authentication

引用

INTERNATIONAL JOURNAL OF COMPUTATIONAL INTELLIGENCE SYSTEMS 2024年第1期17卷 1-12页

作者： Madduluri, Suneetha Kumar, T. Kishore Natl Inst Technol Dept Elect & Commun Engn Warangal 506004 Telangana India

Many societal institutions have increased standards for the efficacy and dependability of identification systems in response to the ever-increasing sophistication of computer technology. Biometric identity systems have mostly supplanted the usage of conventional key and smart card systems in highly secure industries. There has been a dramatic increase in the adoption of biometric authentication systems in the last decade. While many biometric systems have relied on exterior physiological traits like fingerprints, iris scans, palm prints, and faces, very few have investigated the possibility of using interior physiological traits as a biometric. It is challenging for the benefits of unimodal biometric technology to be realized in real-world applications because to its inherent constraints, such as the fact that it only proposes a single piece of information and that environmental factors can impact data verification. While passwords have served their purpose of authenticating users and controlling access, they have also revealed their weaknesses. Concerns about forging or impersonating authentication have prompted the proposal of various Artificial Intelligence (AI)-based solutions. A lot of people are interested in user authentication based on ECGs these days. Unlike other biometrics, an electrocardiogram verifies that the person is actual and alive, making it one of the most trustworthy advanced authentication methods. The P, Q, R, S, and T characteristic points of an ECG signal are responsible for its most salient features. feature selection is a statistical method that can be implemented in a supervised or unsupervised setting via regression or classification. A Convolution Neural Network (CNN) is a type of neural network that extracts and feeds into another neural network, which then classifies those extracted features. The ECG signal serves as input to a feature extraction network. The neural network performs categorization based on the retrieved feature signal

关键词： Electrocardiogram feature extraction, feature vector Associated features Convolution Neural Networks feature processing User identification

来源：评论

学校读者我要写书评

暂无评论

Design of a Traveling Wave Tube Multistage Depressed Collector Based on Back Propagation Neural NetworkRandom Forest Algorithm

引用

IEEE TRANSACTIONS ON PLASMA SCIENCE 2024年第7期52卷 3014-3020页

作者： Xia, Gongao Liu, Wenxin Li, Hongjing Zhou, Dongxing Jiao, Ang Lin, Ruibo Chinese Acad Sci Aerosp Informat Res Inst Beijing 100094 Peoples R China Univ Chinese Acad Sci Sch Elect Elect & Commun Engn Beijing 100049 Peoples R China

The multistage depressed collector (MDC) is a significant component of the traveling wave tube (TWT) that collects the wasted energy of the electron beam after beam-wave interaction. However, more stages of the collectors means the design process of the MDC is more complex and needs more calculations due to more parameter combinations. To simplify the design process and shorten the design time, the number of samples should be reduced during parameter scanning. To solve the above problems, this article combines two powerful algorithms, random forest (RF) and back propagation neural network (BPNN), as the BPNN-RF algorithm. Finally, this novel algorithm successfully forecasts the power collected by a 100-W TWT's three-stage MDC with lower root mean square error (RMSE) than using these two algorithms. The MDC's efficiency of this MDC after optimization is up to 85% when the secondary electron yield (SEY) is 1.2 in the CST studio suite.

关键词： Radio frequency Prediction algorithms Electrons Random forests Backpropagation Electron beams Traveling wave tubes Back propagation neural network (BPNN) feature processing multistage depressed collector (MDC) random forest (RF) traveling wave tube (TWT)

来源：评论

学校读者我要写书评

暂无评论

ViTFSL-Baseline: A Simple Baseline of Vision Transformer Network for Few-Shot Image Classification

引用

IEEE ACCESS 2024年 12卷 11836-11849页

作者： Wang, Guangpeng Wang, Yongxiong Pan, Zhiqun Wang, Xiaoming Zhang, Jiapeng Pan, Jiayun Univ Shanghai Sci & Technol Sch Opt Elect & Comp Engn Shanghai 200093 Peoples R China

Few-shot image classification, whose goal is to generalize to unseen tasks with scarce labeled data, has developed rapidly over the years. However, in traditional few-shot learning methods with CNNs, non-local features and long-rang dependencies of the image may be lost, and this leads to a poor generalization of the trained model. With the advantage of the self-attention mechanism of Transformer, researchers have tried to use vision transformer to improve few-shot learning recently. However, these methods are more complicated and take up a lot of computing resources, and there is no baseline to measure their effectiveness. We propose a new method called ViTFSL-baseline. We take advantage of vision transformer and train our model on all train set without episodic training. Meanwhile, we design a new nearest-neighbor classifier to used for few-shot image classification. Furthermore, in order to narrow the gap between difference of same class, we introduce centroid calibration in classifier after the feature extraction of backbone. We run the experiments on popular benchmarks to show that our method is a simple and effective for few-shot image classification. Our approach could be taken as the baseline upon vision transformer for few-shot learning.

关键词： Deep learning few-shot learning feature processing image classification vision transformer

来源：评论

学校读者我要写书评

暂无评论

PARFormer: Transformer-Based Multi-Task Network for Pedestrian Attribute Recognition

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2024年第1期34卷 411-423页

作者： Fan, Xinwen Zhang, Yukang Lu, Yang Wang, Hanzi Xiamen Univ Minist Educ China Sch Informat Fujian Key Lab Sensing & Comp Smart City Xiamen 361005 Peoples R China Xiamen Univ Minist Educ China Key Lab Multimedia Trusted Percept & Efficient Com Xiamen 361005 Peoples R China

Pedestrian attribute recognition (PAR) has received increasing attention because of its wide application in video surveillance and pedestrian analysis. Extracting robust feature representation is one of the key challenges in this task. The existing methods primarily rely on convolutional neural networks (CNNs) as the backbone network for feature extraction. However, these methods mainly focus on small discriminative regions while ignoring the global perspective. To overcome these limitations, we propose PARFormer, a pure transformer-based multi-task PAR network consisting of four modules. In the feature extraction module, we build a transformer-based strong baseline for feature extraction, which achieves competitive results on several PAR benchmarks compared with the existing CNN-based baseline methods. Since the PAR task is vulnerable to environmental factors, we enhance feature robustness in the feature processing module and propose an effective data augmentation strategy named batch random mask (BRM) block to reinforce the attentive feature learning of random patches. Furthermore, we propose a multi-attribute center loss (MACL) to augment the inter-attribute discriminability of feature representations. As viewpoints can affect some specific attributes, in the viewpoint perception module, we propose a multi-view contrastive loss (MVCL) that enables the network to exploit the viewpoint information. In the attribute recognition module, we alleviate the negative-positive imbalance problem to generate the attribute predictions. These modules interact and jointly learn a highly discriminative feature space and supervise the generation of the final features. Extensive experimental results show that the proposed PARFormer network performs well compared to the state-of-the-art methods on several public datasets, including PETA, RAP, and PA100K. Code will be released at https://***/xwf199/PARFormer.

关键词： Transformers feature extraction Task analysis Visualization Multitasking Image recognition Training pedestrian attribute recognition transformer feature processing viewpoint information

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：