检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Mondal, Sohag Kumar Talukder, Md Simul Hasan Aljaidi, Mohammad Sulaiman, Rejwan Bin Tushar, Md Mohiuddin Sarker Alsuwaylimi, Amjad A. Electrical and Electronic Engineering Khulna University of Engineering and Technology Bangladesh Khulna Khulna Bangladesh E-12/A Agargaon Dhaka1207 Bangladesh Department of Computer Science Faculty of Information Technology Zarqa University Zarqa Jordan School of Computer science and Technology Northumbria University United Kingdom Electrical and Electronic Engineering Bangabandhu Sheikh Mujibur Rahman Science and Technology University Bangladesh Department of Information Technology College of Computing and Information Technology Northern Border University Arar Saudi Arabia

Blood cell classification and counting are vital for the diagnosis of various blood-related diseases, such as anemia, leukemia, lymphoma, and thrombocytopenia. The manual process of blood cell classification and counting is time-consuming, prone to errors, and labor-intensive. Therefore, we have proposed a deep learning (DL)-based automated system for blood cell classification and counting from microscopic blood smear images. We classify a total of nine types of blood cells, including Erythrocyte, Erythroblast, Neutrophil, Basophil, Eosinophil, Lymphocyte, Monocyte, Immature Granulocytes, and Platelet. Several preprocessing steps like image resizing, rescaling, contrast enhancement and augmentation are utilized. To segment the blood cells from the entire microscopic images, we employed the U-Net model. This segmentation technique aids in extracting the region of interest (ROI) by removing complex and noisy background elements. Both pixel-level metrics such as accuracy, precision, and sensitivity, and object-level evaluation metrics like Intersection over Union (IOU) and Dice coefficient are considered to comprehensively evaluate the performance of the U-Net model. The segmentation model achieved impressive performance metrics, including 98.23% accuracy, 98.40% precision, 98.25% sensitivity, 95.97% Intersection over Union (IOU), and 97.92% Dice coefficient. Subsequently, a watershed algorithm is applied to the segmented images to separate overlapped blood cells and extract individual cells. We have proposed a BloodCell-Net approach incorporated with custom light weight convolutional neural network (LWCNN) for classifying individual blood cells into nine types. Comprehensive evaluation of the classifier's performance is conducted using metrics including accuracy, precision, recall, and F1 score. The classifier achieved an average accuracy of 97.10%, precision of 97.19%, recall of 97.01%, and F1 score of 97.10%. A 5-fold cross-validation technique is applied to split t

关键词： Blood

来源：评论

学校读者我要写书评

暂无评论

Energy Efficient Spectrum Sharing and Resource Allocation for 6G Air-Ground Integrated Networks

引用

IEEE Transactions on Network and Service Management 2025年

作者： Huang, Tao Liu, Jingyuan Chang, Zheng Wei, Yao Zhao, Xu Liang, Ying-Chang University of Electronic Science and Technology of China School of Computer Science and Engineering Chengdu611731 China University of Jyväskylä Faculty of Information Technology JyväskyläFIN-40014 Finland Chengdu611731 China Research Institute of China Telecom Guangzhou China

In this paper, we investigate the spectrum sharing and resource allocation scheme for air-ground integrated wireless network which consists of multiple unmanned aerial vehicles (UAVs) and a high altitude platform (HAP). We consider the UAVs are required to provide services or execute certain missions in the area that HAP owns the spectrum and other resources. Correspondingly, we propose an energy efficient spectrum sharing and resource allocation scheme so that the UAVs can flexibly utilize the radio resources within the area without degrading the quality of service (QoS) of the HAP. In the proposed scheme, we jointly optimize pricing of spectrum and transmit power to maximize the utility of both the HAP and UAVs in the considered system in an energy efficient manner. A game theoretic approach is then presented to find the spectrum sharing and resource allocation strategies for both HAP and UAVs and the problem has been addressed via convex optimization. Our extensive simulations demonstrate marked improvements in system utility, spectrum and energy efficiency, and also highlight the effectiveness of the proposed scheme. © 2004-2012 IEEE.

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Approach to Optimal Power Allocation in MIMO-NOMA: Integrating DGNN and Tyrannosaurus Optimization 4

A Hybrid Approach to Optimal Power Allocation in MIMO-NOMA: ...

引用

4th International Conference on Sustainable Expert Systems, ICSES 2024

作者： Agrawal, Anurag Vijay Anuradha, A. Patil, Kranti D. Patil, Vilaskumar Karthikumar, K. Tallapragada, S. V. V. Satyanarayana Bhagwant Institute of Technology Department of Electronics and Communication Engineering Uttar Pradesh Muzaffarnagar251315 India Sri Aditya Degree College Department of Computer Science Andhra Pradesh Bhimavaram534202 India Trinity Academy of Engineering Ghrcem Department of E & Tc Maharashtra Pune411048 India Sharnbasva University Faculty of Engineering and Technology Department of Electronics and Communication Engineering Karnataka Kalaburagi585103 India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Electrical and Electronic Engineering Avadi Tamil Nadu Chennai600062 India School of Engineering Mohan Babu University Department of Electronics and Communication Engineering Andhra Pradesh Tirupati517102 India

ISBN: (纸本)9798331540364

One of the primary technologies necessary for the upcoming sixth generation of communication networks (6G) is massive multi-input multi-output, or MIMO. However, enormous MIMO presents a challenge in reducing energy use and the sum rate demand due to the power use of the several radio frequencies (RF) links in 6G technology. Therefore, this research presented a novel Deformable Graph Neural Network (DGNN) with Tyrannosaurus optimization for the optimal power distribution that maximized the sum rate and energy efficiency of the MIMO NOMA system. The adaptive mechanism of DGNN enables more accurate and efficient power allocation decisions. The integrated Tyrannosaurus optimization ensures the model converging on optimal power allocation strategies that maximize both sum rate and energy efficiency. The efficiency of the recommended model is compared in regarding sum rate and Energy Efficiency (EE) with the some of the recent approaches. The proposed method gained the greater sum rate and EE outperforming all other compared models. © 2024 IEEE.

关键词： Radio links

来源：评论

学校读者我要写书评

暂无评论

SWIPT-enabled Cell-free Millimeter-wave MIMO with Multiple IRSs

SWIPT-enabled Cell-free Millimeter-wave MIMO with Multiple I...

引用

IEEE Globecom Workshops

作者： Tan Do-Duy Nguyen Thi Thanh Van Nguyen Cong Luong Dusit Niyato Faculty of Electrical and Electronics Engineering Ho Chi Minh City University of Technology and Education Vietnam Faculty of Electrical and Electronic Engineering PHENIKAA University Hanoi Vietnam Faculty of Computer Science PHENIKAA University Hanoi Vietnam School of Computer Science and Engineering Nanyang Technological University Singapore

In this paper, we investigate a simultaneous wireless information and power transfer (SWIPT)-enabled cell-free millimeter-wave multiple-input multiple-output (MIMO) system. Therein, multiple base stations (BSs) serve multiple users by using the SWIPT based on power-splitting (PS) protocol. To further improve the system throughput and harvested energy, multiple intelligent reflecting surfaces (IRSs) are deployed. We formulate an optimization problem that aims to optimize the beamformers of the BSs, phase shifts of the IRSs, and PS factors at users to maximize the minimum energy efficiency of the users. The max-min energy efficiency optimization problem is computationally intractable, and we propose an alternating descent algorithm to solve it with low computational complexity. Simulation results demonstrate the effectiveness of the proposed algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Lab-on-a chip electrochemical sensing platform for simultaneous, ultra-sensitive and on-spot detection of 4-aminosalicylic acid and 5-aminosalicylic acid based on synergistic potential of chitosan functionalized MWCNTs supported on Ni doped Bi2S3

引用

Chemosphere 2025年 379卷

作者： Chatterjee, Darshana Silva, S. Ravi P. Tiwari, Ida Department of Chemistry Institute of Science Banaras Hindu University Varanasi211005 India Advanced Technology Institute School of Computer Science and Electronic Engineering Faculty of Engineering and Physical Sciences University of Surrey United Kingdom Institute for Sustainability University of Surrey United Kingdom

Mesalamine or 5-aminosalicylic acid (5-ASA) and its isomer 4-aminosalicylic acid (4-ASA), well known key therapeutic agents used to treat inflammatory bowel diseases (IBDs) can pose toxicity risks upon unregulated consumption. However, their simultaneous real-time detection from physiological fluids like urine remains unexplored. This study presents an innovative electrochemical sensing platform using modified screen-printed electrodes capable of simultaneous detection of both the drugs by harnessing the synergistic potential of a novel nanocomposite comprising chitosan functionalized multi-walled carbon nanotubes and nickel doped bismuth sulphide. Comprehensive optical and microstructural characterization validate the modified sensor platform's morphological characteristics. The sensor was evaluated using CV and DPV, exhibiting notably low detection limits which is of the value 39.559 μM for 5-ASA and 85.21 μM for 4-ASA. Sensitivity was found to be 0.174 μA μM−1cm−2 for the linear dynamic range (LDR) of 50 μM–5750 μM for 5-ASA and 0.139 μA μM−1cm−2 for the linear dynamic range (LDR) of 100 μM–2200 μM for 4-ASA. Moreover, the adaptability of the sensor for integration into hand-held point-of-care devices for practical application has been demonstrated in this paper. Experimental validation using real urine samples underscores the sensor's impressive recovery rate of 98–99.6 % for 5-ASA and 95.12–99.24 % for 4-ASA and its capability of detecting target drugs even when present with typical urinary constituents as interferences. The real-world applicability of this sensing platform is further emphasized by conducting experiments on miniaturized hand-held device thus making it a promising tool for on-the-spot detection, offering substantial potential for future integration into point-of-care diagnostic devices to monitor patients requiring precise medical monitoring. Our approach offers unprecedented real-time identification capabilities of 4-ASA and 5-ASA which has not

关键词： Chitosan

来源：评论

学校读者我要写书评

暂无评论

Face Recognition Based On Fusion Feature Extraction Algorithms

Face Recognition Based On Fusion Feature Extraction Algorith...

引用

Emerging Smart Computing and Informatics (ESCI), Conference on

作者： Ibrahim Mahmood Rashid Albakri Muhammad Imran Ahmad Mohd Nazrin Md Isa Mustafa Zuhaer Nayef Al-Dabagh Faculty of Electronic Engineering Technology Universiti Malaysia Perlis Perlis Malaysia Collage of Agriculture Telafer University Telafer Iraq Department of Computer Science Knowledge University Erbil Iraq

Biometric-based methods have emerged as a highly promising avenue for individual recognition, with face recognition standing out as a key application in image analysis. Facial recognition systems are affected by various image condition factors, including expressions, occlusion, poses, and illuminations. It is crucial to consider reasonable variations in illumination between gallery and probe images when developing face recognition algorithms. In the context of face verification, illumination variation significantly influences the likelihood of misclassification. Existing research lacks a focus on imaging conditions, especially regarding illumination, in face recognition systems. This research aims to identify and design a scheme to address illumination variation in facial recognition. The proposed scheme will undergo analysis and evaluation to assess its accuracy and effectiveness. The image processing in this research is divided into two phases. The first is the preprocessing phase, encompassing auto contrast balancing and noise reduction. The second phase is face processing, which involves illumination normalization, face detection, feature extraction, and face recognition. Two datasets are utilized: the first being the ORL dataset, which includes images with illumination variations, and the second being the Yale database, where images contain various noises.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Web Attack Detection Method Based on DistilBERT and Feature Fusion for Power Micro-Application Server 2

A Web Attack Detection Method Based on DistilBERT and Featur...

引用

2nd International Conference on Advanced electronics, Electrical and Green Energy, AEEGE 2023

作者： Nige, Li Lu, Chen Lei, Zhao Zhenning, Teng Zhiqiang, Wang Yiyang, Shao Xiaolin, Gui State Grid Laboratory of Power Cyber-Security Protection and Monitoring Technology State Grid Smart Grid Research Institute Co.Ltd. Nanjing China Xi'an Jiaotong University Faculty of Electronic and Information Engineering Shaanxi Province Key Laboratory of Computer Network Xi'an China

ISBN: (纸本)9798350322866

The widespread use of power micro-applications has significantly expanded the attack surface of their server side, thus increasing the risk of vulnerability attacks such as SQL injection, XSS, and CRLF injection. To ensure the security of power systems, these attacks must be detected precisely and timely. Therefore, this paper proposes a server-side web attack detection method based on DistilBERT and feature fusion for HTTP and HTTPS requests. The method treats HTTP and HTTPS requests as text data, specifically, extracts deep semantic features using DistilBERT, and fuses them with well-designed empirical features to comprehensively characterize and classify the request. Consequently, anomalous requests can be detected. The experimental results show that the accuracy, precision, recall, and F1 score of the method on HTTP CSIC 2010 and FWAF datasets are close to or higher than 99%. Compared with other methods as far as we known, this method has better performance and efficiency. © 2023 IEEE.

关键词： Cybersecurity

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Recognition Method for Student Classroom Behavior

A Deep Learning Recognition Method for Student Classroom Beh...

引用

Artificial Intelligence technology (ACAIT), Asian Conference on

作者： Ziyuan Lin Jieyi Liu Xinran Wu Yu Zhou Yongkang Wang Pai Lin School of Electronic Engineering Xidian University Xi’an China School of Computer Science and Technology Xidian University Xi’an China Faculty of Integrated Circuit Xidian University Xi’an China

ISBN: (数字)9798331517090

ISBN: (纸本)9798331517106

A deep learning method for recognizing student classroom behavior is proposed, aiming to overcome the limitations of existing methods that rely on single-type features and perform poorly in complex scenarios. The method utilizes the AlphaPose algorithm to extract the skeletal keypoint coordinates, capturing skeletal features of student body movements. Head pose features are derived by calculating head Euler angles from facial skeletal keypoints using the Perspective-n-Point method. Additionally, facial images are detected using the Yunet model, and facial features are extracted following image enhancement with the Contrast Limited Adaptive Histogram Equalization method. By integrating these three types of features, a multi-feature fusion network is constructed for classroom behavior recognition, enabling comprehensive analysis for effective classification of student behaviors. Experimental results indicate that this multi-feature fusion network achieves a recognition accuracy of 93.63%, demonstrating a robust capability in recognizing student classroom behavior.

关键词： Deep learning Histograms Head Accuracy Feature extraction Vectors Classification algorithms Artificial intelligence Image enhancement Facial features

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Image Transfer by Simulation 1

Deep Learning Image Transfer by Simulation

引用

1st IEEE International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and computer engineering, MI-STA 2021

作者： Ammar, Abdussalam Masaud Youssef Ellafi, Amira Bozed, Kenz A. Zerek, Amer R. Faculty of Electronic Technology Tripoli Libya Benghazi University Faculty of Information Technology Computer System Design Dept. Benghazi Libya Zawia University Faculty of Engineering Ee Department Zawia Libya

ISBN: (纸本)9781665418560

The image classification is a classical problem of image processing, computer vision and machine learning fields. In this paper we study the image classification using deep learning with Neural style transfer that has been a high risk application for deep learning, make attention from and advertising the effectiveness to both the academic prisons and the general public. However, we have found by removal experiments that optimizing an image in the way neural style transfer does, we can even factor out the deepness (multiple layers of exchange linear and nonlinear transformations) all together and have neural style transfer working to a certain range. We present the VGG-19 Artistic Type Neural Algorithm, which can convert and recombine the image quality and style of natural images. This algorithm allows us to produce new images of high perceptual quality that combine the content of an arbitrary photograph with the appearance of numerous well known portrait. Our results provide new insights into the deep image representations learned by Convolutional Neural Networks (CNN) and demonstrate their potential for high level image synthesis and manipulation. © 2021 IEEE.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

High-Fidelity Document Stain Removal via A Large-Scale Real-World Dataset and A Memory-Augmented Transformer

arXiv

引用

arXiv 2024年

作者： Li, Mingxian Sun, Hao Lei, Yingtie Zhang, Xiaofeng Dong, Yihang Zhou, Yilin Li, Zimeng Chen, Xuhang School of Computer Science and Engineering Huizhou University China Faculty of Science and Technology University of Macau China School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences China School of Electronic and Communication Engineering Shenzhen Polytechnic University China

Document images are often degraded by various stains, significantly impacting their readability and hindering downstream applications such as document digitization and analysis. The absence of a comprehensive stained document dataset has limited the effectiveness of existing document enhancement methods in removing stains while preserving fine-grained details. To address this challenge, we construct StainDoc, the first large-scale, high-resolution (2145 × 2245) dataset specifically designed for document stain removal. StainDoc comprises over 5,000 pairs of stained and clean document images across multiple scenes. This dataset encompasses a diverse range of stain types, severities, and document backgrounds, facilitating robust training and evaluation of document stain removal algorithms. Furthermore, we propose StainRestorer, a Transformer-based document stain removal approach. StainRestorer employs a memory-augmented Transformer architecture that captures hierarchical stain representations at part, instance, and semantic levels via the DocMemory module. The Stain Removal Transformer (SRTransformer) leverages these feature representations through a dual attention mechanism: an enhanced spatial attention with an expanded receptive field, and a channel attention captures channel-wise feature importance. This combination enables precise stain removal while preserving document content integrity. Extensive experiments demonstrate StainRestorer’s superior performance over state-of-the-art methods on the StainDoc dataset and its variants StainDoc Mark and StainDoc Seal, establishing a new benchmark for document stain removal. Our work highlights the potential of memory-augmented Transformers for this task and contributes a valuable dataset to advance future research. Copyright © 2024, The Authors. All rights reserved.

关键词： Large datasets

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：