检索结果-内蒙古大学图书馆

Deep Neural Network Based on Sparse Auto-Encoder for Road Extraction 19th

Deep Neural Network Based on Sparse Auto-Encoder for Road Ex...

19th EAI International Conference on Heterogeneous Networking for Quality, Reliability, Security and Robustness, QShine 2023

作者： Liu, Sheng Chang, Shuxiao Cao, Ting Li, Xinyue Department of Computer Science and Engineering Xi’an University of Technology Shaanxi Xi’an China

ISBN: (纸本)9783031651250

Road extraction from aerial image has realistic significance for GIS data updating. In view of the complexity challenging for acquiring road information, this paper proposes supervised model that combines Convolutional Neural Network (CNN) with Sparse Auto-Encoder (SAE) to cope with the road extraction task. First, the road features are extracted from the amount of non-annotated data using SAE model that aim to train the road features using CNN principle with implementing convolution and pooling to reduce model complexity. Second, the encoder network completes the operation, and after the deep pooling and deconvolution operations, the intermediate features are extracted by the decoder network and sampled back to the input image of the same size on the map. Third, the soft-max classifier categorizes images into roads and non-roads. Finally, the experiments verify that the proposed method outperforms the traditional methods and could achieve the satisfy result. © ICST Institute for computer sciences, Social Informatics and Telecommunications engineering 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Evaluation of Classifiers Based on Neural Network for Handwritten Digit Recognition

Evaluation of Classifiers Based on Neural Network for Handwr...

引用

2024 International Conference on Advances in Computing Research on science engineering and technology, ACROSET 2024

作者： Sehta, Neha Saraf, Krupi Vinchi, Aditya Acropolis Institute of Technology and Research Department of Computer Science and Engineering Indore India

ISBN: (纸本)9798350388800

Handwritten Digit Recognition (HDR) refers to the computer' s ability to identify and categorize hand written digits from various sources into 10 predetermined categories, ranging from 0 to 9. The issue of digit recognition is significant and crucial. Due to variations in size, thickness, location, and orientation of manually written digits, several challenges need to be addressed in order to evaluate the problem of recognizing handwritten digits. The diversity and variety in the composition techniques of different persons also impact the pattern and appearance of the digits. It refers to the approach of recognizing and organizing written numbers. It offers a diverse array of uses, such as automated bank checks, postal addresses, and tax paperwork, among others. The objective of this research is to employ a classification system to accurately identify handwritten numerals. The consequences of the most widely employed deep learning algorithms, such as multilayered perceptron network, in conjunction with convolution neural networks, are being discussed. Comparison of five neural network classifier shows that accuracy up to 99% can be achieved only in few epochs of training with default configuration of hyper-parameters. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Optimizing classifier performance using PCA-FLANN: a fast and reliable approach

引用

Iran Journal of computer science 2025年 1-14页

作者： Jena, Monalisa Behera, Ranjan Kumar Dehuri, Satchidananda Cho, Sung-Bae Department of Computer Science Yonsei University Seoul 03722 South Korea Department of Computer Science & Engineering Indian Institute of Information Technology Jharkhand Ranchi India Department of Computer Science Fakir Mohan University Odisha Balasore India

This work introduces PCA-FLANN, an innovative hybrid model combining principal component analysis (PCA) with functional link artificial neural network (FLANN) to achieve efficient non-linear dimensionality reduction and robust classification. The approach is motivated by the need to optimize feature selection, recognizing that not all features equally contribute to achieving optimal accuracy. PCA is employed to extract the most relevant features from datasets, retaining essential information while significantly reducing dimensionality, and FLANN processes these reduced features to handle non-linear correlations effectively. To assess its performance, PCA-FLANN is evaluated on six widely-used real-world datasets from the KEEL repository and compared against hybrid models such as PCA-RBFN, PCA-SVM, PCA-RF, PCA-MLP, and standalone FLANN. The results reveal that PCA-FLANN outperforms these models in accuracy, with improvements of 8.5%, 2.94%, 9.09%, 11.1%, and 9.69%, respectively. Additionally, the model demonstrates enhanced computational efficiency and reduced complexity, establishing PCA-FLANN as a powerful and precise solution for classification tasks involving dimensionality reduction. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： Classification Feature extraction FLANN Neural network PCA

来源：评论

学校读者我要写书评

暂无评论

Ramp Progressive Secret Image Sharing using Ensemble of Simple Methods 14

Ramp Progressive Secret Image Sharing using Ensemble of Simp...

引用

14th International Conference on computer and Knowledge engineering, ICCKE 2024

作者： Mokhtari, Atieh Taheri, Mohammad Department of Computer Science and Engineering and Information Technology Shiraz University Shiraz Iran

ISBN: (纸本)9798331511272

This paper reviews the limitations of traditional Secret Image Sharing (SIS) methods and recent advancements such as PSIS, RPSIS, and SSIS, and presents an optimized and practical combinatorial method for secret image sharing by ensemble of simple methods. In this approach, a color or grayscale image is divided into n shares, each of which is individually meaningless. The image is reconstructed through XOR operations between corresponding pixels in different shares. The main innovation of this method is the introduction of two thresholds, k1 and k2. This scheme is characterized by parameters (k1, k2, n), where having fewer than k1 shares reveals no information about the original image, while between k1 and k2 shares, the image quality progressively improves and reaches near-complete reconstruction. Ultimately, with k2 or more shares, the image can be recovered in its full resolution. This method allows for customization of the image quality improvement rate between k1 and k2 and addresses the limitations of previous designs regarding the selection of k1, k2, and n values. Its flexibility in providing suitable models based on specific user needs makes this method appropriate for various applications in secret image sharing. ©2024 IEEE.

关键词： Progressive Recovery Progressive Secret Image Sharing Ramp Secret Image Sharing Secret Image Sharing Secret Sharing Visual Cryptography

来源：评论

学校读者我要写书评

暂无评论

Exploring Contextual Appropriateness of Spectral Indices for Remote Sensing-Based Land Coverage Classification Using Machine Learning

Exploring Contextual Appropriateness of Spectral Indices for...

引用

2024 International Conference on Artificial Intelligence and Emerging technology, Global AI Summit 2024

作者： Singh, Abhishek Sahil National Institute of Technology Delhi Department of Computer Science and Engineering Delhi India

ISBN: (纸本)9798350379716

Spectral indices obtained from satellite images are helpful for remote sensing, particularly for land cover classification (LCC). Therefore, the purpose of this paper will be to evaluate the performance of the three regularly used spectral indices, including Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and Normalized Difference Built-up Index (NDBI), in disaggregating three classes of land cover, i.e., vegetation, water, and built-up areas. The work uses sentinel-2 generated satellite data from Google Earth Engine and Machine Learning classifiers to consider the contextual appropriateness of Spectral Indices for LCC of Remote sensing. The research proves that all indices are beneficial and non-beneficial in recognizing multiple forms of land covering;in other words, NDVI is more precise in identifying the vegetation-based land types than the other two indices. Contrary to this, NDWI is best suited to water-based land coverage, while NDBI is best suited to built-up areas-based land coverage. The study builds on the remote sensing domain by reviewing several spectral indices and determining their suitability for use by the LCC sub-domain. © 2024 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Transfer learning based lightweight model for classification of Alzheimer's disease using brain MR images

Transfer learning based lightweight model for classification...

引用

2024 OPJU International technology Conference on Smart Computing for Innovation and Advancement in Industry 4.0, OTCON 2024

作者： Dash, Prasanta Kumar Sisodia, Dilip Singh National Institute of Technology Raipur Department of Computer Science and Engineering Raipur India

ISBN: (纸本)9798350373783

Alzheimer's disease (AD) is one of the prevailing, degenerative, and progressive brain disease which is the primary cause of dementia all over the world. It steadily worsens by affecting the memory, behavioral patterns, and reasoning of an individual. Traditional manual assessment of AD progression using patient's brain MRI scans by radiologist could lead to certain errors with consequential impacts for patient. So, automated end-to-end frameworks are used to identify various AD stages at an early stage. Deep neural networks are widely used to detect AD using MR images. However, the scarcity of MR images for training the Deep Learning models is a significant problem for AD identification. To resolve this issue, transfer learning techniques are used with different pre-train models such as VGGNet, ResNet, DenseNet, and InceptionNet to classify AD using MR images. However, transfer learning models generally take more training parameters with more memory space. Hence, it is difficult to deploy these models in resource constraint environments such as medical applications. To overcome the above problem, we designed a lightweight deep neural network using a pre-trained MobileNetV1 with an optimized architecture that employs depth-wise separable convolution and finetune the model for AD classification. The proposed model was evaluated against models like VGG16, DenseNet, AlexNet, and ResNet50 in existing literatures using evaluation measures such as recall, precision, F1 score, and accuracy. The proposed model outperformed other cutting-edge models used in this study, achieving a precision of 98%, accuracy of 97.8%, recall of 95%, and F1 score of 96%. © 2024 IEEE.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Detection of Malicious URL Websites using Machine Learning Models 4

Detection of Malicious URL Websites using Machine Learning M...

引用

4th International Conference on Electronics and Sustainable Communication Systems, ICESC 2023

作者： Reddy, Karri Narendra Venkatesh, Kolapalli Naga Prem Jacob, T. Sathyabama Institute of Science and Technology Computer Science and Engineering Chennai India

ISBN: (纸本)9798350300093

In recent years, web applications have become increasingly vulnerable to hacking, with an estimated occurrence of 32,590 attacks every day. However, many web developers and website owners are unprepared to identify and prevent these attacks. Hackers utilize various techniques, including phishing websites to gain unauthorized access or compromise authentic web programmers. This research study examines the web application security and the frequency of attacks on such systems. This study focuses on the most common types of attacks and suggests effective detection methods for preventing them. Recently, the secure coding methodologies and machine learning algorithms are used to detect and block unauthorized access and phishing *** almost all the existing research works, a web application is developed to test several website URLs. The findings suggest that the model is capable of detecting malicious web application attacks. Furthermore, the model compares the performance of several machine learning algorithms for identifying phishing website links and detecting with the best model. © 2023 IEEE.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

YOLOv7-ConvTrans: Hybrid Vision Transformer with YOLOv7 for Underwater Object Detection 4

YOLOv7-ConvTrans: Hybrid Vision Transformer with YOLOv7 for ...

引用

4th IEEE International Conference on Mobile Networks and Wireless Communications, ICMNWC 2024

作者： Delsey, M.J. Bibal Benifa, J.V. Indian Institute of Information Technology Department of Computer Science and Engineering Kottayam India

ISBN: (数字)9798331528348

ISBN: (纸本)9798350352931

Underwater object detection plays a significant role in marine exploration activities such as ecological monitoring, conservation of undersea ecosystems, and underwater robotics. In contrast to detection in the atmosphere on land, underwater detection is challenging due to color distortion and attenuation, which render the detection process more difficult and imprecise. To address these issues, this paper proposes an improved detection framework, YOLOv7-ConvTrans. The YOLOv7 architecture is optimized by integrating a convolution vision transformer in the neck region to enhance the detection accuracy of small and occluded objects. The proposed method reduces redundancy, computational cost, and recognition time to extract significant feature maps from the imagery. In the experimental assessment on the URPC2019 dataset, a mAP of 80.39 and on the Brackish dataset, 92.61, indicate high detection accuracy. Finally, our proposed approach provides a valuable framework for designing underwater vision systems that can be utilized in various scientific and exploratory domains. © 2024 IEEE.

关键词： Machine vision

来源：评论

学校读者我要写书评

暂无评论

Identification and Analysis of Neural Disorders Based on Hyperactivity and Spectrum Disorder with Functional MRI 8th

Identification and Analysis of Neural Disorders Based on Hyp...

引用

8th International Conference on computers, Management and Mathematical sciences, ICCM 2022

作者： Kodukula, Sai Siddharth Varanasi, Usha Bala Boddikura, Soumya Talluri, Pavani Kalepu, Bala Sai Computer Science and Engineering Anil Neerukonda Institute of Technology and Sciences Visakhapatnam India

ISBN: (纸本)9783031511660

A collection of conditions known as neurodevelopmental disorders affect a number of factors, including mood, learning capacity, self-control, and memory. The most frequent and prominent are attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD). The current diagnosis methods are time-consuming, human-mediated, and unreliable. This study focuses on the prediction of ADHD and ASD, and a comparative analysis is done using neural networks, support vector machine, gradient boosting, and logistic regression models to determine which performs well for the resting-state functional MRI (fMRI) data. During this procedure, time-series signals were collected from the brain region’s voxels and utilized to create functional connectivity characteristics. Our models were evaluated on various metrics, and we observed that a multi-layer perceptron model performs well. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Brain

来源：评论

学校读者我要写书评

暂无评论

Targeted Influence Maximization Using Centrality and Diminishing Weight

Targeted Influence Maximization Using Centrality and Diminis...

引用

2024 IEEE Region 10 Conference, TENCON 2024

作者： Baruah, Khanjan Changmai Tripathi, Rohit Indian Institute of Information Technology Department of Computer Science and Engineering Guwahati India

ISBN: (纸本)9798350350821

Social Networks are an integral part of our daily life. These networks are tools for sharing information among different users. In such a network, the most influential users become the source for disseminating information over the network. Hence, social network analysis was done to understand and identify the importance of a node in information diffusion. These influential nodes help to achieve maximum influence throughout the network. However, in certain cases, only a subset of nodes are needed to be influenced. This is called targeted influence maximization. In this paper, we use different centrality measures and diminishing weights to identify influential nodes and study the influence of these nodes on the target nodes. Our simulation shows that the seed selection scheme using centrality and diminishing weights activates more target nodes in a network than the existing schemes. © 2024 IEEE.

关键词： Centrality Measures Independent Cascade Model Social Network Analysis Target Influence Maximization

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：