检索结果-内蒙古大学图书馆

6th International Conference on Frontier Technologies of Information and Computer, ICFTIC 2024

作者： Zhang, Xuebiao Chen, Jia Zhang, Lidong Zhai, Yingdong School of Railway Transportation Shanghai Institute of Technology Shanghai China Shanghai Inesa Network Co. Ltd. Department of Railway Transportation Shanghai China Technology Center Shanghai Shentong Metro Group Co. Ltd. Shanghai China

ISBN: (纸本)9798331541750

With the rapid development of rail transit, station management faces increasing complexity and an explosive growth in data volume. To address the challenges posed by the high dimensionality and diversity of passenger flow data, this study proposes an enhanced clustering model that integrates KAE (kernel method-based autoencoder, KAE) for dimensionality reduction with a hybrid clustering approach combining DBSCAN and K-Means. The KAE is utilized to reduce the dimensionality of the high-dimensional data while retaining key features, which improves the effectiveness of the subsequent clustering process. In this model, DBSCAN is applied to identify high-density regions in the data, which are then used to initialize the K-Means algorithm. This ensures that the initial centroids are positioned in areas with high data density. The proposed approach combines the capability of DBSCAN to detect clusters of arbitrary shapes based on density with the ability of K-Means to refine cluster centroids for better convergence. The use of KAE further enhances the clustering performance by optimizing the feature space, making it more suitable for clustering. Experiments conducted on both rail transit and general datasets demonstrate that the proposed model achieves superior clustering performance compared to traditional methods. The results show improved classification accuracy and robustness, confirming the model's strong generalization capability across different types of data. This method provides an efficient and scalable solution for the classification of rail transit stations and offers new insights for managing and analyzing high-dimensional data in various applications. © 2024 IEEE.

关键词： component dbscan hybrid clustering algorithm k-means kernel method autoencoder urban rail transit

来源：评论

学校读者我要写书评

暂无评论

Medical Image Recognition Using a Novel Neural Network Construction Controlled by a kernel method Encoder

引用

IEEE ACCESS 2024年 12卷 146807-146817页

作者： Li, Pengzhi Pei, Yan Li, Jianqiang Xie, Haihua Univ Aizu Grad Sch Comp Sci & Engn Aizu Wakamatsu 9658580 Japan Beijing Univ Technol Sch Software Engn Beijing 100124 Peoples R China Changsha Vocat & Tech Coll Sch Informat Engn & Business Changsha 410019 Peoples R China

The recognition and diagnosis of medical images is one of the important topics in artificial intelligence, such as colonic polyp and cataract classification. The appearance, shape, and size of pathological features in medical images significantly influence the accuracy of identification. Improving recognition accuracy is crucial due to the difficulty in distinguishing pathological features accurately. This paper presents a medical image recognition method utilizing a neural network controlled by a kernel method encoder, which presents the originality of this research. Medical images are enhanced to extract various features. Using the kernel method encoder, a high-dimensional distinguishable feature map is generated. This mitigates the problem of insufficient features in a single image. Image classification is achieved by optimizing kernel method encoder parameters and training the enhanced neural network classification model. We conducted experiments on colonography computed tomography images and colour fundus images. Initially, we conducted classification experiments on the colonography computed tomography image dataset. The experimental results demonstrate that the proposed model improves classification accuracy and shows good performance. Subsequently, we conducted the same classification experiment on colour fundus images using the proposed method. The experimental results indicate that this method improves the classification accuracy. The accuracy of this method on two medical image datasets is 92.1% and 97.4%, respectively. The proposed method performs well across both medical image datasets. This further demonstrates the robustness of the multi-feature and enhanced neural network model classification method.

关键词： Feature extraction kernel Biomedical imaging Medical diagnostic imaging Cataracts Computed tomography Image recognition Image classification Colonic polyps Mathematical models image classification kernel method autoencoder medical image 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

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