检索结果-内蒙古大学图书馆

Visual impairment prevention by early detection of diabetic retinopathy based on stacked auto-encoder

SCIENTIFIC REPORTS 2025年第1期15卷 1-31页

作者： Almas, Shagufta Wahid, Fazli Ali, Sikandar Alkhyyat, Ahmed Ullah, Kamran Khan, Jawad Lee, Youngmoon Univ Haripur Dept Informat Technol Haripur 22620 Pakistan Univ Derby Sch Comp Coll Sci & Engn Derby DE22 3AW England Islamic Univ Coll Tech Engn Najaf 54001 Iraq Univ Haripur Dept Biol Haripur 22620 Pakistan Gachon Univ Sch Comp Seongnam 13120 South Korea Hanyang Univ Dept Robot Ansan 15588 South Korea Islamic Univ Al Diwaniyah Coll Tech Engn Dept Comp Tech Engn Al Diwaniyah 58001 Iraq Islamic Univ Babylon Coll Tech Engn Dept Comp Tech Engn Babylon 51002 Iraq

Diabetic retinopathy (DR) presents a significant concern among diabetic patients, often leading to vision impairment or blindness if left untreated. Traditional diagnosis methods are prone to human error, necessitating accurate alternatives. While various computer-aided systems have been developed to assist in DR detection, there remains a need for accurate and efficient methods to classify its stages. In this study, we propose a novel approach utilizing enhanced stacked auto-encoders for the detection and classification of DR stages. The classification is performed across one healthy (normal) stage and four DR stages: mild, moderate, severe, and proliferative. Unlike traditional CNN approaches, our method offers improved reliability by reducing time complexity, minimizing errors, and enhancing noise reduction. Leveraging a comprehensive dataset from KAGGLE containing 35,126 retinal fundus images representing one healthy (normal) stage and four DR stages, our proposed model demonstrates superior accuracy compared to existing deep learning algorithms. Data augmentation techniques address class imbalance, while SAEs facilitate accurate classification through layer-wise unsupervised pre-training and supervised fine-tuning. We evaluate our model's performance using rigorous quantitative measures, including accuracy, recall, precision, and F1-score, highlighting its effectiveness in early disease diagnosis and prevention of blindness. Experimental results across different training/testing ratios (50:50, 60:40, 70:30, and 75:25) showcase the model's robustness. The highest accuracy achieved during training was 93%, while testing accuracy reached 88% on a training/testing ratio of 75:25. Comparative analysis underscores the model's superiority over existing methods, positioning it as a promising tool for early-stage DR detection and blindness prevention.

关键词： Diabetic retinopathy Disability Deep learning stacked auto-encoder Dropout

来源：评论

学校读者我要写书评

暂无评论

stacked auto-encoder-Based Fault Location in Distribution Network

引用

IEEE ACCESS 2020年 8卷 28043-28053页

作者： Luo, Guomin Tan, Yingjie Li, Meng Cheng, Mengxiao Liu, Yanmei He, Jinghan Beijing Jiaotong Univ Sch Elect Engn Beijing 100044 Peoples R China

Accurate fault location is crucial for finding and clearing faults in distribution networks. It can help to reduce the loss of power failure and ensure safe and stable operation. Without efficient observation points and the lack of phase angle information, traditional fault location methods of distribution networks usually have large errors. As the application of fiPMUs in the distribution network becomes more and more common, voltage and current waveforms, as well as phase angle can be recorded and provide more information for accurate fault location. In this paper, an intelligent location method is proposed to pinpoint the fault location based on the information of fiPMUs which are properly allocated. The fault section is firstly determined by comparing the zero-sequence current waveforms on both sides of the fault section. Then, a Stack auto-encoder (SAE) is modeled to provide an end-to-end solution to pinpoint fault point with the voltage and current phasors. Finally, the performance of the proposed method is tested by a simulated distribution network on the platform of PSCAD. The results show that the proposed method is effective in locating faults and can withstand the effects of transition resistance, fault type, and noise. Compared to another popular method, the proposed method shows better location accuracy.

关键词： Fault location stacked auto-encoder distribution network

来源：评论

学校读者我要写书评

暂无评论

stacked auto-encoder Based Fault Location in VSC-HVDC

引用

IEEE ACCESS 2018年 6卷 33216-33224页

作者： Luo, Guomin Yao, Changyuan Liu, Yinglin Tan, Yingjie He, Jinghan Wang, Kai Beijing Jiaotong Univ Sch Elect Engn Beijing 100044 Peoples R China CloudMinds Technol Inc Beijing 100102 Peoples R China

This paper presents an end-to-end approach for locating faults on high-voltage dc (HVDC) transmission lines. Different from traditional methods which rely on communications between different measuring units or feature extraction of post fault transients, the proposed algorithm takes the raw data of locally detected traveling current surges as the only-input and outputs the fault locations directly. Especially, the stacked auto-encoder (SAE) is utilized to model the relationship between fault currents and fault locations. The SAE-based method is performed in time domain and tested with a simulated HVDC transmission line modeled in PSCAD/EMTDC. The simulation results show that this method is effective in locating faulted points and robust against attenuation, overlapping of traveling surges, and various ground resistances.

关键词： VSC-HVDC fault location stacked auto-encoder deep learning

来源：评论

学校读者我要写书评

暂无评论

stacked auto-encoder-Based Transients Recognition in VSC-HVDC

引用

IEEE ACCESS 2020年 8卷 14223-14233页

作者： Luo, Guomin Cheng, Mengxiao Sun, Hao Li, Meng Tan, Yingjie He, Jinghan Zhang, Heng Beijing Jiaotong Univ Sch Elect Engn Beijing 100044 Peoples R China State Grid Beijing Maintenance Co Beijing 100069 Peoples R China

For overhead long-distance high voltage direct current (HVDC) transmission lines, transients are produced due to complicated field conditions and lightning activities. To ensure reliable operation of protection devices, accurate recognition of faults and disturbances is quite critical. The most popular recognition methods include threshold-based ones which require the proper setting of the threshold value, and classifier-based ones that need suitable feature extractions. These methods depend heavily on the experience of engineers or experts and are ineffective in dealing with the variation of system parameters. In this paper, a transient recognition method based on stack auto-encoder (SAE) is proposed to characterize different transients and to avoid human interferences. A symmetrical +/- 500kv HVDC system is modeled to illustrate the performance of the proposed method. The effect of some factors, such as noises and conductors, are discussed and compared. The simulation results demonstrate that the proposed SAE-based recognition has excellent potential in transient recognition of practical HVDC systems.

关键词： stacked auto-encoder VSC-HVDC transient recognition

来源：评论

学校读者我要写书评

暂无评论

Bearing condition monitoring via an unsupervised and enhanced stacked auto-encoder

引用

JOURNAL OF THE BRAZILIAN SOCIETY OF MECHANICAL SCIENCES AND ENGINEERING 2024年第6期46卷 1-12页

作者： Xu, Fan Hao, Zhenyu Zhou, Chao Deng, Yaling Wuhan Univ Technol Sch Mech & Elect Engn Wuhan 430070 Hubei Peoples R China CASIC Inst Laser & Photont Engn Wuhan 430075 Hubei Peoples R China Wuhan Univ Technol Sch Management Wuhan 430070 Hubei Peoples R China Wuhan Univ Technol Res Inst Digital Governance & Management Decis Inn Wuhan 430070 Hubei Peoples R China

Supervised deep learning models have been widely used in the construction of bearing health indicators (HIs) for performance degradation. Such models require a series of model improvement combinations, which are dependent on manual experience, to extract HIs. Additionally, noise leads to a large, extracted health indicator value that easily exceeds the preset threshold, causing an incorrect evaluation of health status. Therefore, this paper proposes an improved and unsupervised stacked auto-encoder without an output label layer and with local weighted regression and smoothing scatterplot (Lowess). That is, a stacked auto-encoder adds a Lowess filter after each hidden layer to perform multiple de-noising operations. We used encoders and decoders at several hidden layers within a stacked noise auto-encoder (though only with an output layer) to extract HIs and directly eliminate the noise from the original bearing vibration. Hence, our proposed method reduces the modeling complexity. The results show that our proposed method outperforms others of its type, such as stacked auto-encoders, stacked de-noising auto-encoders, deep belief neural networks, root-mean-square kurtosis, empirical mode decomposition, singular-value decomposition-K-medoids, and self-organizing map networks. Moreover, by using the monotonicity evaluating indicator, the extracted HI curve was found to be smoother than in other models and the probability of erroneous judgments for bearing performance degradation assessment was reduced.

关键词： Deep learning Health indicator Bearing stacked auto-encoder Local weighted regression Smoothing scatterplot

来源：评论

学校读者我要写书评

暂无评论

Using stacked auto-encoder and Bi-Directional LSTM For Batch Process Quality Prediction

引用

JOURNAL OF CHEMICAL ENGINEERING OF JAPAN 2021年第4期54卷 144-151页

作者： Qi, Jiakang Luo, Na East China Univ Sci & Technol Key Lab Adv Control & Optimizat Chem Proc Minist Educ Shanghai 200237 Peoples R China

Batch process quality prediction has broad application prospects in manufacturing and chemical industries. However, during the final quality prediction of a batch process, the final target values may be related to the whole process track of the batch reaction. Thus, the final quality prediction problem embraces complex high-dimensional input and simple low-dimensional output, which also means a serious size mismatch between input data and predictive values. Motivated by these difficulties, a hybrid prediction model is proposed, which combines the advantages of stacked auto-encoder (SAE) and bi-directional long short-term memory (BLSTM) for the final quality prediction of a batch process. The feature extraction ability of SAE is used to obtain the low-dimensional features of historical process data along the time direction. Then, the validity of the framework was verified by taking penicillin fermentation as an example.

关键词： Batch Process Quality Prediction stacked auto-encoder Long Short-Term Memory

来源：评论

学校读者我要写书评

暂无评论

Wear indicator construction for rolling bearings based on an enhanced and unsupervised stacked auto-encoder

引用

SOFT COMPUTING 2023年第15-16期28卷 8835页

作者： Zeng, Wenhui Yu, Lisha Xu, Fan Huang, Zhelin Zhou, Shengwen Guo, Shunsheng Du, Baigang Wuhan Univ Technol Sch Mech & Elect Engn Wuhan Peoples R China Hong Kong Polytech Univ Sch Design Hong Kong Peoples R China Shenzhen Univ Coll Econ Dept Stat Shenzhen Peoples R China

The degradation state of a bearing can be monitored effectively by using a wear indicator (WI). A WI curve having smoothness and monotonicity can lay a good foundation for predicting the remaining useful life of the bearing. Most traditional models for bearing WI construction, such as time-frequency indicators and signal decomposition, are complicated;for example, some WI construction models need several models to select from, and their fusion depends on the manual experience of engineers. For example, single and mixed traditional time-frequency indicators, such as the root mean square (RMS), Kurtosis, multiple time-frequency domain fusion. However, the mentioned-above time-frequency domain indicators are difficult to adaptively reflect the operating status of the equipment when the operating conditions of the mechanical system change. Some signal decomposition models are combined with other models and rely on manual experience to extract WI, such as the selection of effective intrinsic mode function components function, parameter setting of empirical mode decomposition and ensemble empirical mode decomposition model, etc. Deep learning models, such as stacked auto encoder (SAE) and convolution neural network, have been widely used in bearing health monitoring and WI construction, because its powerful learning and feature extraction capabilities of multiple hidden layer structures. But these deep learning models are designed to use output labels. Particularly when the data volume is large, it requires manpower, material resources, and experienced engineers to label the data, or it is difficult to label and distinguish the categories of data samples. Therefore, to solve these problems and eliminate the need for manual labor, such as labeling data and selecting models for fusion, we propose using SAE without an output label layer to extract WI from original signals directly. However, an extracted WI curve without good monotonicity (Mon) will result in a poor remaining

关键词： Roller bearings Deep learning stacked auto-encoder Wear indicator Sine function

来源：评论

学校读者我要写书评

暂无评论

Detecting 3D Points of Interest Using Multiple Features and stacked auto-encoder

引用

IEEE TRANSACTIONS ON VISUALIZATION AND COMPUTER GRAPHICS 2019年第8期25卷 2583-2596页

作者： Shu, Zhenyu Xin, Shiqing Xu, Xin Liu, Ligang Kavan, Ladislav Zhejiang Univ Sch Comp & Data Engn Ningbo Inst Technol Ningbo 315000 Zhejiang Peoples R China Shandong Univ Sch Comp Sci & Technol Jinan 250000 Shandong Peoples R China Ningbo Univ Sch Informat Sci & Engn Ningbo 315000 Zhejiang Peoples R China Univ Sci & Technol China Graph & Geometr Comp Lab Sch Math Sci Hefei 230000 Anhui Peoples R China Univ Utah Sch Comp Salt Lake City UT 84112 USA

Considering the fact that points of interest on 3D shapes can be discriminated from a geometric perspective, it is reasonable to map the geometric signature of a point p to a probability value encoding to what degree p is a point of interest, especially for a specific class of 3D shapes. Based on the observation, we propose a three-phase algorithm for learning and predicting points of interest on 3D shapes by using multiple feature descriptors. Our algorithm requires two separate deep neural networks (stacked auto-encoders) to accomplish the task. During the first phase, we predict the membership of the given 3D shape according to a set of geometric descriptors using a deep neural network. After that, we train the other deep neural network to predict a probability distribution defined on the surface representing the possibility of a point being a point of interest. Finally, we use a manifold clustering technique to extract a set of points of interest as the output. Experimental results show superior detection performance of the proposed method over the previous state-of-the-art approaches.

关键词： 3D shapes point of interest multiple features stacked auto-encoder

来源：评论

学校读者我要写书评

暂无评论

Ensemble stacked auto-encoder Classification on LIDAR Remote Sensing Images

引用

JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING 2018年第4期46卷 597-604页

作者： Li, Dawei Zhang, Ruifang North Univ China Sch Comp & Control Engn Taiyuan 030051 Shanxi Peoples R China Jin Ind Grp Co Ltd Taiyuan 030027 Shanxi Peoples R China

Light detection and ranging system (LIDAR) can obtain diverse remote sensing datasets which contains different land cover information. The datasets offer vital and significant features for land cover classification. As a new and effective deep learning algorithm, stacked auto-encoders (SAE) consists of multiple auto-encoders in which the code of each auto-encoder is the input of the successive one. The classification precision is closely related to hidden layers, and the number of samples in fine-tuning step also affects classification results. In this paper we study the classifiers based on different number of samples and hidden layers. According to appropriate parameters, we promote SAE with adaptive boosting ensemble strategy to build new classification method. Two tests which are based on LIDAR datasets are implemented. The experiment results prove that the fusion of deep learning and ensemble learning is effective to LIDAR remote sensing images. The proposed method is robust to similar scenes classification. The overall accuracy increases 6% compared with bagging method on test 1.

关键词： stacked auto-encoder Ensemble learning LIDAR Deep learning

来源：评论

学校读者我要写书评

暂无评论

Spatial-Spectral Jointed stacked auto-encoder-Based Deep Learning for Oil Slick Extraction from Hyperspectral Images

引用

JOURNAL OF THE INDIAN SOCIETY OF REMOTE SENSING 2019年第12期47卷 1989-1997页

作者： Liu, Bingxin Zhang, Qiang Li, Ying Chang, Wen Zhou, Manrui Dalian Maritime Univ Nav Coll Dalian 116026 Peoples R China TAL Educ Grp Beijing 100080 Peoples R China China Waterborne Transport Res Inst 8 Xitucheng Rd Beijing 100088 Peoples R China Shenyang Normal Univ Educ Coll Shenyang 110034 Liaoning Peoples R China

Hyperspectral remote sensing provides an outstanding tool in oil slick detection and classification, for its advantages in abundant spectral information. Many classification methods have been proposed and tested for oil spill extraction using hyperspectral images. However, the deep learning method was hardly researched to classify oil slicks using hyperspectral images. In this work, we proposed a spatial-spectral jointed stacked auto-encoder (SSAE) to extract and classify oil slicks on the sea surface. The traditional machine learning methods, support vector machine, back-propagation neural network and stacked auto-encoder were also adopted. The experimental results reveal that our proposed SSAE model can remarkably outperform the other models, especially for the thick oil films. The results of this work could provide an alternative method to extract oil slicks on hyperspectral remote sensing images.

关键词： Hyperspectral remote sensing Oil spill stacked auto-encoder

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：