检索结果-内蒙古大学图书馆

Leveraging swin transformer with ensemble of deep learning model for cervical cancer screening using colposcopy images

引用

SCIENTIFIC REPORTS 2025年第1期15卷 1-22页

作者： Himabindu, D. Dakshayani Lydia, E. Laxmi Rajesh, M. V. Ahmed, Mohammed Altaf Ishak, Mohamad Khairi VNR Vignana Jyothi Inst Engn & Technol Dept Informat Technol Hyderabad India Siddhartha Acad Higher Educ VR Siddhartha Engn Coll A Dept Informat Technol Vijayawada India Aditya Univ Dept Informat Technol Surampalem India Prince Sattam Bin Abdulaziz Univ Coll Comp Engn & Sci Dept Comp Engn Al Kharj 11942 Saudi Arabia Ajman Univ Coll Engn & Informat Technol Dept Elect & Comp Engn Ajman U Arab Emirates

Cervical cancer (CC) is the leading cancer, which mainly affects women worldwide. It generally occurs from abnormal cell evolution in the cervix and a vital functional structure in the uterus. The importance of timely recognition cannot be overstated, which has led to various screening methods such as colposcopy, Human Papillomavirus (HPV) testing, and Pap smears to identify potential threats and enable early intervention. Early detection during the precancerous phase is crucial, as it provides an opportunity for effective treatment. The diagnosis and screening of CC depend on colposcopy and cytology models. Deep learning (DL) is an appropriate technique in computer vision, which has developed as a latent solution to increase the efficiency and accuracy of CC screening when equated to conventional clinical inspection models that are vulnerable to human error. This study presents a Leveraging Swin Transformer with an Ensemble of Deep Learning Model for Cervical Cancer Screening (LSTEDL-CCS) technique for colposcopy images. The presented LSTEDL-CCS technique aims to detect and classify CC on colposcopy images. Initially, the wiener filtering (WF) model is used for image pre-processing. Next, the swin transformer (ST) network is utilized for feature extraction. For the cancer detection process, the ensemble learning method is performed by employing three models, namely autoencoder (AE), bidirectional gated recurrent unit (BiGRU), and deep belief network (DBN). Finally, the hyperparameter tuning of the DL techniques is performed using the pelican optimization algorithm (POA). A comprehensive experimental analysis is conducted, and the results are evaluated under diverse metrics. The performance validation of the LSTEDL-CCS methodology portrayed a superior accuracy value of 99.44% over existing models.

关键词： Deep learning Autoencoder Cervical cancer Colposcopy image Bidirectional gated recurrent unit pelican optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

POC-net: pelican optimization-based convolutional neural network for recognizing large pose variation from video

引用

NEURAL COMPUTING & APPLICATIONS 2023年第34期35卷 24091-24107页

作者： Jayabharathi, P. Suresh, A. SRM Inst Sci & Technol Sch Comp Dept Networking & Commun Chennai 603203 Tamil Nadu India SRM Inst Sci & Technol Fac Engn & Technol Sch Comp Dept Networking & Commun Chennai 603203 Tamil Nadu India

Nowadays, face recognition using video surveillance systems becomes one of the active research topics in security domains. Security plays a significant role in everyday life for secure and sustainable developments of smart cities. The conventional techniques provide efficient recognition results only when the faces are captured with complete face images. However, they suffer to handle large pose variation images extracted from video sequences. Therefore, to deal with this issue, this paper specially designed a multidimensional face recognition model to recognize faces under multiple pose variations and angles. Three face video databases, namely facesurv database, IARPA Janus benchmark database and McGill database, are utilized for experimental evaluation. The videos of these three databases are converted into number of image frames through background subtraction process. From the image frames, the large pose variation images with different angles are identified and selected to process further. The video recorded under dynamic environment conditions diminishes recognition performance, so the image frames are processed through several preprocessing pipelines. The preprocessed images are then fed into the proposed optimal mask region-based convolutional neural network with modified short-term memory (OMRCNN-MBiLSTM) model, which learns the facial patterns present in the images more efficiently. The feature vectors learned by the proposed classifier are matched with the input face database to determine the identity of the person. With the ability to handle multiview and large pose variations, the proposed model accurately recognizes faces. The simulation result manifests the superiority of proposed model over other existing methods.

关键词： Face recognition Pose variation Mask region-based convolutional neural network Bidirectional long short-term memory pelican optimization algorithm

来源：评论

学校读者我要写书评

暂无评论

pelican Whale optimization Enabled Deep Learning Framework for Video Steganography Using Arnold Transform-Based Embedding

引用

INTERNATIONAL JOURNAL OF PATTERN RECOGNITION AND ARTIFICIAL INTELLIGENCE 2024年第2期38卷 2359026-2359026页

作者： Suresh, G. Manikandan, G. Bhuvaneswari, G. Shanthakumar, P. Kings Engn Coll Dept Comp Sci & Engn Sriperumbudur 602117 Tamil Nadu India Kings Engn Coll Dept Informat Technol Sriperumbudur 602117 Tamil Nadu India Saveetha Engn Coll Dept Comp Sci & Engn Chennai 602105 Tamil Nadu India Nehru Inst Technol Dept Comp Sci & Engn Coimbatore 641105 Tamil Nadu India

Steganography refers to hiding a secret message from various sources, such as images, videos, audio and so on. The advantage of steganography is to avoid data hacking in transmission medium during the transmission of information sources. Video steganography is superior to image steganography since the videos can hide a substantial quantity of secret messages more than the image. Hence, this research introduced the video stereography technique, Arnold Transform with SqueezeNet-based pelican Whale optimization algorithm (AT+SqueezeNet_PWOA), for concealing the secret image on the video. To hide the secret image on the video, the proposed method follows three steps: key frame and feature extraction, pixel prediction and embedding. The extraction of the key frame process is carried out by the Structural Similarity Index Measure (SSIM), and then the neighborhood features and convolutional neural network (CNN) features are extracted from the frame to improve the robustness of the embedding process. Moreover, the pixel prediction is completed by the SqueezeNet model, wherein the learning factors are tuned by the PWOA. In addition, the embedding process is completed by applying the Arnold transform on the predicted pixel, and the transformed regions are combined with the secret image using the embedding function. Likewise, the extraction process extracts the secret image from the embedded video by substituting the predicted pixel and Arnold transform on the embedded video. The proposed method is used to hide chunks of secret data in the form of video sequences and it improves the performance. The Arnold transform used in this work provides security by encrypting the data. The use of SqueezeNet makes the proposed model a simple design and this reduces the computational time. Thus, the AT+SqeezeNet_PWOA attained better correlation coefficient (CC), peak signal-to-noise ratio (PSNR) and mean square error (MSE) of 0.908, 48.66 and 0.001 dB with the Gaussian noise.

关键词： SqueezeNet Arnold transform pelican optimization algorithm Whale optimization algorithm convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Secure Virtual Machine Migration and Host Overload Detection Using Modified pelican optimization with Variable Load Mean Function

引用

JOURNAL OF CIRCUITS SYSTEMS AND COMPUTERS 2024年第14期33卷 2450255-2450255页

作者： Parthasarathy, S. SRM Valliammai Engn Coll Dept Comp Sci & Engn Kattankulathur 603203 Tamil Nadu India

Low-resource utilization and high-energy consumption have become progressively protuberant issues in cloud data centers. Virtual Machine (VM) migration is the key objective to resolve this issue. Moreover, extreme VM migration might empower Service-Level Agreement (SLA) violations. Few works are considered for optimizing throughput and energy consumption. An efficient VM migration must consider different parameters like network communication overhead, migration overhead, resource utilization, energy consumption and quality of service which is a multi-objective issue. Hence, in this paper, a Modified pelican optimization-based Variable Load Mean Function (MPO-VLMF)-based host overload detection is presented and security enhancement is developed. The main motive of this study is to achieve host overload detection and security enhancement. To obtain host overload detection, the variable load mean function is developed. In this mean function, the weight parameter is selected by considering Modified pelican optimization (MPO). The Levy flight (LF) is considered for enhancing the updating process of pelican optimization (PO). To enhance the security of the system, the Digital Signature-based Encryption (DSE) is developed. Based on this proposed approach, security and host overload detection are obtained. The proposed technique is implemented and evaluated by performance measures. It is compared with the conventional approaches to justify the performance of the system.

关键词： Host overload detection secure virtual machine migration levy flight function pelican optimization algorithm variable load mean function

来源：评论

学校读者我要写书评

暂无评论

Harnessing feature pruning with optimal deep learning based DDoS cyberattack detection on IoT environment

引用

Scientific Reports 2025年第1期15卷 1-15页

作者： Yang, Eunmok Jeong, Sooyong Seo, Changho Department of Financial Information Security Kookmin University Seoul 02707 South Korea Department of Convergence Science Kongju National University Gongju 32588 South Korea Basic Science Research Institution Kongju National University Gongju 32588 South Korea

The swift development of the Internet of Things (IoT) devices has created a pressing need for effective cybersecurity measures. They are vulnerable to different cyber threats that can compromise the functionality and security of urban systems. Distributed Denial of Service (DDoS) attacks are among IoT networks’ most challenging and destructive cyber threats. With the rapid growth in IoT devices and users, the vulnerability of IoT devices to such attacks has enhanced significantly, making DDoS attacks a predominant threat. This work introduces several approaches for effectively detecting IoT-based DDoS threats. Classical machine learning (ML) techniques mostly face difficulty in managing real-world traffic characteristics effectually, making them less appropriate for detecting DDoS attacks. In contrast, Artificial Intelligence (AI)-based methods have proven more effective in detecting cyber-attacks than conventional approaches. This manuscript proposes an effective Feature Pruning with Optimal Deep Learning-based DDoS Attack Detection (FPODL-DDoSAD) technique in the IoT framework. The FPODL-DDoSAD technique initially uses a min-max scalar for the data scaling into the standard layout. Besides, the feature pruning process is performed using an improved pelican optimization algorithm (IPOA), which enables the choice of an optimal subset of features. Meanwhile, DDoS attacks are recognized using a sparse denoising autoencoder (SDAE) model. Furthermore, the parameter tuning of the SDAE classifier is accomplished by utilizing the Fish Migration Optimizer (FMO) technique. The experimental values of the FPODL-DDoSAD approach are assessed on the benchmark BoT-IoT dataset. The comparison study of the FPODL-DDoSAD method demonstrates a superior accuracy value of 99.80% over existing techniques. © The Author(s) 2025.

关键词： Cyberattack detection Fish migration optimizer Min–max scalar pelican optimization algorithm Sparse denoising autoencoder

来源：评论

学校读者我要写书评

暂无评论

Fuzzy-based pelican optimization for battery management and energy scheduling in solar aided electric vehicle

ENGINEERING RESEARCH EXPRESS

引用

ENGINEERING RESEARCH EXPRESS 2024年第2期6卷

作者： Sharma, Pankaj Mishra, Ambarisha NIT Patna Elect Engn Dept Patna India

Electric vehicles (EVs) are considered an essential mode of transportation due to their advantages, non-polluting and noise-free characteristics. The batteries are considered the primary power source for electric vehicles, ensuring energy supply and thus improving the user experience. However, designing charging stations and uncontrolled charge scheduling are the barriers evaluated in charging stations. To solve these issues, a fuzzy-based pelican optimization algorithm (POA) is proposed in this work. Set of rules are generated in the fuzzy for optimal energy management inn charging station. POA schedules the EVs in the charging station based on arrival time and SOC to reduce the running cost. Demand in the charging station is based on the state of charge (SOC) of the EVs arriving at the charging station. Solar power with battery and ultracapacitor is used as an input source for charging stations. The energy management in the system is based on fuzzy to satisfy the load demand from PV and energy storage devices. This proposed method is implemented in the Simulink tool to evaluate the performance. This simulation result show the effectiveness of the proposed with optimal scheduling and energy management. By this method running cost are reduced compared to the existing algorithms.

关键词： charging station electric vehicle energy management fuzzy logic controller pelican optimization algorithm scheduling state of charge

来源：评论

学校读者我要写书评

暂无评论

Research on tidal energy prediction method based on improved time-varying filter-empirical mode decomposition and confluent double-stream neural network

引用

OCEAN ENGINEERING 2024年第Part3期312卷

作者： Huang, Yi Li, Guohui Xian Univ Posts & Telecommun Sch Elect Engn Xian 710121 Shaanxi Peoples R China

Tidal energy is a new type of clean energy, the development and utilization of it has great scientific research potential, practical application value and broad development prospect. Because the generation process of tidal energy involves many complex ocean dynamic factors, and the method of predicting tidal energy is relatively simple, it is a meaningful work to research its prediction. Aiming at the nonlinear and nonstationary characteristics of tidal energy, propose a tidal energy prediction method based on improved time-varying filter- empirical mode decomposition with pelican optimization algorithm (POA-TVF-EMD) and confluent double- stream neural network (CNN-LSTM). Firstly, propose a sum of refined composite multiscale dispersion entropy and Spearman correlation coefficient as the fitness function of the search process, and use POA-TVF-EMD to decompose tidal energy into a string of intrinsic mode functions. Secondly, use CNN-LSTM to predict each component, and reconstruct the component prediction result to get the initial prediction result. Then, calculate the difference between the original signal and the initial prediction result to get the error result, bring the error result back to CNN-LSTM for prediction to get the error prediction result, and take the average of the error result and the error prediction result to get the error correction result. Finally, reconstruct the initial prediction result and the error correction result to get the final prediction result. Take tidal energy data in Texas, Caribbean and Washington as case study, and design ten prediction methods for comparative experiment. Take tidal energy data of Washington as an example, coefficient of determination, root mean square error, mean absolute error and mean absolute percentage error of the prediction results of proposed prediction method are 0.99873, 0.10695, 0.08189 and 0.00965 respectively. The experimental result shows that proposed prediction method is superior to all other comp

关键词： Tidal energy Mode decomposition pelican optimization algorithm Neural network Prediction Error correction

来源：评论

学校读者我要写书评

暂无评论

Attack classification in network intrusion detection system based on optimization strategy and deep learning methodology

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2024年第31期83卷 75533-75555页

作者： Ramu, Ch. Kodanda Rao, T. Srinivasa Rao, E. Uma Shankar GITAM Dept CSE Visakhapatnam 530045 Andhra Pradesh India

In recent times, network-based applications have rapidly grown in the field of information and communication technology(ICT), which enables individuals and organizations to connect and share their sensitive information seamlessly. The security of these network-based applications is imperative to avoid cyber-attacks during the exchange of sensitive information. The identification of anomalies in network events can be highly challenging due to the complex nature of traffic flows. To solve the challenges, network intrusion detection system (NIDS) technology is used;any network can profit from this system because it can monitor traffic and detect any irregularities. The existing NIDS systems driven by machine learning models do not provide sufficient ability to handle heterogeneous data and realize performance degradation while detecting some types of attacks. Therefore, this paper proposes an innovative meta-heuristic optimization and deep learning-based methodology for improving the performance of NIDS systems. Initially, the raw captured traffic data is fed into the pre-processing phase to attain data standardization and data balancing. Further, an extended pelican optimization algorithm (Ex-Pel) is employed to select the set of features from the pre-processed data optimally. Finally, the Self-Attention Assisted Weighted Auto Encoder (SAttn_WAE) is executed to detect the attacks precisely through the set of optimal features. The execution of the proposed methodology is carried out in the Python platform, and performance evaluation is done using accuracy, precision, recall, FAR, and F1-score metrics. The proposed model achieved an accuracy of 99.23%, precision of 99.78%, FAR of 0.67%, recall of 99.13%, and F1-score of 99.32%, which is comparatively better than existing models.

关键词： Cyber attacks Network intrusion detection system Meta-heuristic optimization Deep learning pelican optimization algorithm Self-attention Weighted auto encoder

来源：评论

学校读者我要写书评

暂无评论

An optimized intrusion detection model for wireless sensor networks based on MLP-CatBoost algorithm

引用

MULTIMEDIA TOOLS AND APPLICATIONS 2024年第25期83卷 66725-66755页

作者： Francis, E. Geo Sheeja, S. Karpagam Acad Higher Educ Dept Comp Sci Coimbatore Tamil Nadu India

A wireless Sensor Network (WSN) is made up of many sensor nodes which gather and transmit data to a central location. The limited resources of the nodes create significant security challenges when deploying and communicating WSNs. The detection of unauthorized access is a crucial aspect of enhancing the security measures of WSNs. The utilization of network intrusion detection systems (IDS) has become an essential aspect of any communication network, as they offer valuable services to the network. Several studies in the field of machine learning have been conducted to explore the potential of utilizing this technology for intrusion detection in WSNs, yielding promising outcomes. These efforts still need to be more precise and efficient against network traffic unbalanced data issues. The paper presents a new model for detecting intrusion attacks that utilize a hybrid multilayer perceptron (MLP) and CatBoost classifier, as well as feature selection techniques. The proposed approach aims for good performance in identifying different forms of threats. The system performs data preprocessing on various datasets and reduces the dataset size using a feature selection algorithm. pelican optimization algorithm (POA) has been proposed for tuning the hyper-parameters of the classifier designs and selecting the relevant features from the dataset. The CSE-CIC-IDS2018, AWID, and UNSW-NB15 databases reutilized for conducting performance evaluations on the proposed framework. The tests included accuracy, precision, recall, FAR, DR and complexity time. The proposed model has a low FPR and high accuracy in binary classification, as shown.

关键词： Wireless sensor networks Feature selection Machine learning Hybrid model pelican optimization algorithm CatBoost Multilayer perceptron

来源：评论

学校读者我要写书评

暂无评论

A Two-Tiered Bidirectional Atrous Spatial Pyramid Pooling-Based Semantic Segmentation Model for Landslide Classification Using Remote Sensing Images

引用

IEEE ACCESS 2024年 12卷 181316-181331页

作者： Nalinipriya, G. Lydia, E. Laxmi Alshenaifi, Reem Kavuri, Radhika Ishak, Mohamad Khairi Saveetha Engn Coll Dept Informat Technol Chennai 602105 India Siddhartha Acad Higher Educ Velagapudi Ramakrishna Siddhartha Engn Coll Dept Informat Technol Vijayawada 520007 India Majmaah Univ Coll Comp & Informat Sci Dept Informat Technol Al Majmaah 11952 Saudi Arabia Chaitanya Bharathi Inst Technol AI & DS Dept Hyderabad 500075 India Ajman Univ Coll Engn & Informat Technol Dept Elect & Comp Engn Ajman U Arab Emirates

Effective landslide representation from great spatial resolution images is significant in numerous applications. Many research works and techniques have been advertised. Still, these methods are very challenging to relate in real time since they depend on remotely sensing landslides from a solitary sensor with an exact spatial resolution. Precisely identifying landslides over a vast region with intricate background entities is difficult. Machine Learning (ML) and Deep Learning (DL) have attained extraordinary performance in classifying images utilizing remotely sensed images from numerous platforms. Moreover, techniques built within DL architectures tend to implement encoder-decoder network structures, where constant convolutions effortlessly strain out numerous landslide features. This study develops a Bidirectional Atrous Spatial Pyramid Pooling-Based Semantic Segmentation and Classification Model (BASPP-SSCM) technique for landslide Remote Sensing Images. The main goal of the BASPP-SSCM technique is to segment and classify the landslide areas. In the preprocessing stage, the BASPP-SSCM model employs an adaptive Wiener filtering (AWF) technique to eliminate the noise. Next, for the semantic segmentation method, the BASPP-SSCM technique utilizes the DeepLabV3 method with the backbone of the ConvNeXtLarge model for determining the landslide region. Furthermore, the CapsNet model is utilized for the feature extraction process. Besides, the Rigdelet neural network (RNN) technique is employed for the landslide classification process. At last, the pelican optimization algorithm (POA) methodology is implemented to fine-tune the parameters involved in the RNN model. A wide range of investigations is performed to highlight the superiority of the BASPP-SSCM method using a benchmark dataset. The performance validation of the BASPP-SSCM method underscored a superior accuracy value of 98.23% of other existing approaches.

关键词： Terrain factors Feature extraction Remote sensing Semantic segmentation Adaptation models Accuracy Image recognition Computational modeling Spatial resolution Reliability landslide remote sensing images pelican optimization algorithm feature extraction atrous spatial pyramid pooling

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

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： 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：