检索结果-内蒙古大学图书馆

7th International Conference on Inventive Computation Technologies, ICICT 2024

作者： Sudhesh, K.M. Akhila, T. Thangavel, Senthil Kumar Shanmugam, Selvanayaki Kolandapalayam Amrita Vishwa Vidhyapeetham Amrita School of Computing Department of Computer Science And Engineering Coimbatore India Ashland University Department of Mathematics And Computer Science OH United States

ISBN: (纸本)9798350359299

Agriculture is essential to the global economy and food security. Insect and bird attacks have now been found in recent research that could cause crop loss. Crops are prone to bird attacks throughout the early stages of crop production and are particularly sensitive from the milky stages up to maturation. Observing such birds, which are harmful to agriculture, is an expensive, time-consuming, and possibly error-prone operation. Furthermore, a vast area necessitates a huge labour force for bird scaring. The concept of automated monitoring through computer vision holds significant appeal. This is because it has the potential to facilitate the collection of comprehensive data on a much larger scale than is currently feasible. In this work, we leverage effective characteristics from images of 10 different hazardous birds, Especially in the realm of identification we employing transfer-learned deep learning models which holds promise. The resemblance of various bird features, as well as the difficulty of distinguishing them at a glimpse, has become a significant challenge. We used more than 1000 images of harmful birds from various species for model training. In experiments on two different transfer learned deep CNN models, the xceptionNet model showed the best performance with a test accuracy of 94% when compared to the relatively low memory model VGG19, which had a test accuracy of 80%.The analysis also revealed that the xception model performed better in terms of accuracy, precision, recall, and F1-score. © 2024 IEEE.

关键词： Birds

来源：评论

学校读者我要写书评

暂无评论

CAC-Net: A fire detection method based on residual feature extraction and a Self-Attention mechanism

CAC-Net: A fire detection method based on residual feature e...

引用

2024 International Conference on Signal Processing and Communication Security, ICSPCS 2024

作者： Jiang, Xuefeng Fang, Xianjin Xu, Liuquan School of Safety Science and Engineering Anhui University of Science and Technology Huainan232001 China School of Computer Science and Engineering Anhui University of Science and Technology Huainan232001 China

ISBN: (纸本)9781510681699

Early fire Due to the small target and the occurrence of early scene complexity, the existing fire detection methods make it very easy to miss the detection of false detection phenomena. To address the above problems, this study designs CAC-Net, a fire detection method based on residual feature extraction and a self-attention mechanism. The method utilizes an efficient two-step residual feature extraction module to finely tune the receptive field and enhance the extraction of complex features as well as small object features. Effective Pyramid Aggregation Module AF accurately aggregates task-related information. The CGD module uses group convolution and techniques to efficiently arrange convolution filters for efficient flow and integration of information between different feature map channels, achieving lower false detections. In this paper, we conduct detailed and rich experiments on a live-collected dataset and achieve 96.4% precision and 96.1% recall. The experimental results demonstrate that the CAC-Net model significantly reduces the misdetection and omission of fires and has efficient research value in the task of early fire detection in complex scenarios. © 2024 SPIE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Attention Mechanisms in U-Net Variants for Breast Cancer Image Segmentation 5

Comparative Analysis of Attention Mechanisms in U-Net Varian...

引用

5th IEEE Global Conference for Advancement in Technology, GCAT 2024

作者： George, Geomol Anusuya, S. Simats School of Engineering Department of Electronics and Communication Chennai India Simats School of Engineering Department of Computer Science Chennai India

ISBN: (数字)9798350376685

ISBN: (纸本)9798350376685

Accurate segmentation of breast cancer images is crucial for effective detection and treatment. This work evaluates the efficacy of various Attention U-Net models with different attention mechanisms, including SE (Squeeze-and-Excitation) blocks and Swish activation functions, for breast cancer image segmentation. We assessed these models using several metrics: IoU, precision, recall, F1 score, MAE, dice coefficient, and specificity. The SE-Swish U-Net variant achieved the highest performance metrics, with a mean IoU of 77.71%, precision of 78.6%, recall of 73.5%, F1 score of 75.9%, dice coefficient of 75.91%, and specificity of 96.70%. The SE-Swish model also demonstrated superior accuracy and efficiency compared to other variants, though it required longer training and execution times. Qualitative analysis supported the SE-Swish model's ability to handle complex lesion borders and low-contrast areas. This work highlights how advanced attention mechanisms in U-Net architectures can enhance breast cancer image segmentation. © 2024 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Exploring the Potential of Deep Learning Algorithms in Medical Image Processing: A Comprehensive Analysis 5th

Exploring the Potential of Deep Learning Algorithms in Medic...

引用

5th International Conference on Communication and Intelligent Systems, ICCIS 2023

作者： Pal, Ganesh Prasad Pal, Raju Department of Computer Science and Engineering Jaypee Institute of Information Technology Noida India Department of Computer Science and Engineering School of Information and Communication Technology Gautam Buddha University Greater Noida India

ISBN: (纸本)9789819720781

In this study, we comprehensively examine the potential of deep learning algorithms in the domain of medical image processing. Through a systematic analysis of existing literature, we explore the applications, methodologies, and outcomes of these algorithms across diverse medical specialties. Our analysis underscores the transformative impact of convolutional neural networks and transfer learning techniques in enhancing diagnostic accuracy, image segmentation, and disease detection. We discuss challenges such as data variability, ethics, and interpretability while emphasizing the importance of interdisciplinary collaboration. By synthesizing findings and identifying future directions, we highlight the promising role of deep learning in revolutionizing healthcare diagnostics and treatment planning. After conducting a thorough analysis of various literature and empirical studies, we have evaluated the capabilities and drawbacks of deep learning models in managing different medical imaging modalities such as X-rays, MRIs, and CT scans. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

A Novel Metaheuristic Algorithm: The Team Competition and Cooperation Optimization Algorithm

引用

computers, Materials & Continua 2022年第11期73卷 2879-2896页

作者： Tao Wu Xinyu Wu Jingjue Chen Xi Chen Amir Homayoon Ashrafzadeh School of Computer Science Chengdu University of Information TechnologyChengdu610225China School of Computer Science and Engineering Southwest Minzu UniversityChengdu610041China CSIT Department School of ScienceRMIT UniversityMelbourne3058Australia

Metaheuristic algorithm is a generalization of heuristic algorithm that can be applied to almost all optimization *** optimization problems,metaheuristic algorithm is one of the methods to find its optimal solution or approximate solution under limited *** of the existing metaheuristic algorithms are designed for serial ***,existing algorithms still have a lot of room for improvement in convergence speed,robustness,and *** address these issues,this paper proposes an easily parallelizable metaheuristic optimization algorithm called team competition and cooperation optimization(TCCO)inspired by the process of human team cooperation and *** proposed algorithm attempts to mathematically model human team cooperation and competition to promote the optimization process and find an approximate solution as close as possible to the optimal solution under limited *** order to evaluate the performance of the proposed algorithm,this paper compares the solution accuracy and convergence speed of the TCCO algorithm with the Grasshopper Optimization Algorithm(GOA),Seagull Optimization Algorithm(SOA),Whale Optimization Algorithm(WOA)and Sparrow Search Algorithm(SSA).Experiment results of 30 test functions commonly used in the optimization field indicate that,compared with these current advanced metaheuristic algorithms,TCCO has strong competitiveness in both solution accuracy and convergence speed.

关键词： Optimization metaheuristic algorithm

来源：评论

学校读者我要写书评

暂无评论

Analysis of Sensing Matrix Design for Under-Sampling MRI 9

Analysis of Sensing Matrix Design for Under-Sampling MRI

引用

9th International Conference on Communication and Information Processing, ICCIP 2023

作者： Andryani, Nur afny catur Oey, Endra Laurent, Velix Arvino Nasya, Alexandra Husin, Husna Sarirah Computer Science Department School of Computer Science Bina Nusantara University Indonesia Computer Engineering Department Faculty of Engineering Bina Nusantara University Indonesia Taylor's University Malaysia

ISBN: (纸本)9798400708909

MRI currently is the most powerful medical diagnostic tomography system. It provides high resolution with high detail medical image. Its magnetization sensing is also free from radiation impact hence safe for the patient. However, the data acquisition process takes quite a long time with high-rate sampling. The sensing patient should stay still in the MRI sensing device for 20 to 40 minutes. It can affect claustrophobia for some patients. The high-rate sampling will also linearly affect the cost, which is considerably high. Hence, the research on enabling under-sampling MRI imaging has been raised. The main objective of under-sampling MRI Imaging is reducing the scanning time by under-sampling the data acquisition while at the same time the resulting qualified reconstructed image can be maintained. Compressive Sensing (CS) framework is one of the approaches that can be utilized on solving projection with less measurement. It can recover signals which have sparsity property. The signal is sparse if the signal being sensed has low information rate. Thus, the framework of Compressive Sensing matches under-sampling MRI Imaging. The sparsity requirement is closely related to the design of sensing matrix and dictionary on the Compressive Sensing framework. This paper presents analysis of two proposed sensing matrix designs for Under-sampling MRI Imaging: square sensing matrix and sorting-based sensing matrix designs. The proposed sensing matrix design is intended to appropriately sample the data measurement on certain sampling rate in frequency domain to maintain the quality of the reconstructed MRI image closely to the ground truth (full sampling MRI Imaging). The simulations show that both proposed sensing matrix designs are able to conduct under-sampling MRI Imaging for at least at 40% sampling rate with average of PSNR 27.72 and 33.07 respectively subject to the ground truth. © 2023 Copyright held by the owner/author(s)

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

Face Recognition From Video Using Enhanced Social Collie Optimization-based Deep Convolutional Neural Network 14

Face Recognition From Video Using Enhanced Social Collie Opt...

引用

14th International Conference on Computing Communication and Networking Technologies, ICCCNT 2023

作者： Musale, Jitendra Chandrakant Kumar Singh, Anuj Shirke, Swati Shri Jagdishprasad Jhabarmal Tibrewala University Computer Science and Engineering Rajasthan India School of Computing Science and Engineering Galgotias University Computer Science and Engineering Greater Noida India Mit Adt University Computer Science and Engineering Maharashtra Pune India

ISBN: (纸本)9798350335095

The face recognition system is a process which is part of computer vision as a key feature of video surveillance systems. In a face recognition system stands on two main pillar which are object recognition and authentication of object or person who present in a frame with the help of distinctive picture generation like i.e. Cameras. In this field of computer Vision many researchers work on detection and verification of faces from frames in the subsequent part of video. It is a complex and complicated process to recognize accurate faces by the use of artificial vision algorithms. In machine learning main important concept is Convolutional neural networks (CNNs) in real time system which has video process input for face recognition systems. For such complex problem effective approaches to work on by solution as CNNs have excelled capability to handle, with significant results. Which also including process based voice recognition, face localization, face recognition and picture frame selection and categorization. Recently all the top performing techniques which work on the preprocessed Labelled Face dataset for the Real Time system for convolutional neural networks (CNNs).So here in this research use of deep convolutional neural network (optimized Deep CNN) to getting generated precise outcome of face images from the input video by the Hybrid weighted texture pattern descriptor (HWTP). The help of deep CNN techniques with improved tuned featured extraction and introduce novel ideas about enhanced social collie optimization (ESCO), which result in better solutions with different combinations of strategies. The finest parameters are used to recognise the face of an individual is determined. The proposed model by deep convolutional neural network achieves 91.30% accuracy, 95.07% precision, 95.078% recall and 95.078% f-measures. © 2023 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

SEFormer:A Lightweight CNN-Transformer Based on Separable Multiscale Depthwise Convolution and Efficient Self-Attention for Rotating Machinery Fault Diagnosis

引用

computers, Materials & Continua 2025年第1期82卷 1417-1437页

作者： Hongxing Wang Xilai Ju Hua Zhu Huafeng Li State Key Laboratory of Mechanics and Control for Aerospace Structures Nanjing University of Aeronautics and AstronauticsNanjing210016China School of Computer Science and Engineering Nanyang Technological UniversitySingapore639798Singapore

Traditional data-driven fault diagnosis methods depend on expert experience to manually extract effective fault features of signals,which has certain ***,deep learning techniques have gained prominence as a central focus of research in the field of fault diagnosis by strong fault feature extraction ability and end-to-end fault diagnosis ***,utilizing the respective advantages of convolution neural network(CNN)and Transformer in local and global feature extraction,research on cooperating the two have demonstrated promise in the field of fault ***,the cross-channel convolution mechanism in CNN and the self-attention calculations in Transformer contribute to excessive complexity in the cooperative *** complexity results in high computational costs and limited industrial *** tackle the above challenges,this paper proposes a lightweight CNN-Transformer named as SEFormer for rotating machinery fault ***,a separable multiscale depthwise convolution block is designed to extract and integrate multiscale feature information from different channel dimensions of vibration ***,an efficient self-attention block is developed to capture critical fine-grained features of the signal from a global ***,experimental results on the planetary gearbox dataset and themotor roller bearing dataset prove that the proposed framework can balance the advantages of robustness,generalization and lightweight compared to recent state-of-the-art fault diagnosis models based on CNN and *** study presents a feasible strategy for developing a lightweight rotating machinery fault diagnosis framework aimed at economical deployment.

关键词： CNN-Transformer separable multiscale depthwise convolution efficient self-attention fault diagnosis

来源：评论

学校读者我要写书评

暂无评论

AST-Based and Token-Based Neural Networks for Source Code Classification: A Comparative Performance Analysis 15

AST-Based and Token-Based Neural Networks for Source Code Cl...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Parvathy, R. Thushara, Mg Amrita School of Computing Department of Computer Science & Engineering Amrita vishwa Vidyapeetham Amritapuri India Amrita School of Computing Department of Computer Science & Applications Amrita vishwa Vidyapeetham Amritapuri India

ISBN: (纸本)9798350370249

Programming languages function as a medium for expressing instructions in the context of computer program creation. A prevalent convention in programming is categorizing source code into fragments, a strategy that enhances understanding and simplifies the upkeep process. This research study presents a comparative examination of two distinct models utilized to classify source code. The first model employs a neural network approach that heavily relies on Syntax Trees (ASTs). In contrast, the second model is built on a neural network framework that utilizes the tokenization process. The AST-based model employs a methodology that entails decomposing extensive ASTs into smaller clusters of statement trees. These clusters are subsequently transformed into vectors by capturing the lexical and syntactical characteristics of the statements. The naturalness of certain statements is captured by employing a bidirectional Recurrent Neural Network (RNN) model, which utilizes a collection of statement vectors. In contrast, the token-based paradigm diverges from ASTs and emphasizes tokens obtained through tokenization. The tokens undergo preprocessing and vectorization before inputting into a bidirectional RNN model to evaluate the naturalness of the statements. To choose the optimal model, the accuracy of the two models are compared. © 2024 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Advancing Brain State Classification through Electroencephalogram (EEG) Analysis: Exploring Diverse Models for Improved Classification 7

Advancing Brain State Classification through Electroencephal...

引用

7th International Conference on Contemporary Computing and Informatics, IC3I 2024

作者： Chauhan, Naman Tomar, Riya Atri, Pranshul Chauhan, Chhavi Prakash, Abhijith Vikram, Aditya Amrita Vishwa Vidyapeetham Amrita School of Computing Department of Computer Science and Engineering Tamil Nadu Chennai India University of Petroleum and Energy Studies School of Computer Science Uttarakhand Dehradun India Bauhaus-Universität Weimar Department of Computer Science Weimar Germany

ISBN: (纸本)9798350350067

The combination of electroencephalography data and machine learning technologies provides a promising path for secure and transparent solutions to various problems, including authentication. The outcome of the current work provides an alternative to passwords and biometrics: the use of EEG as a method for identifying brain states such as stress and emotion. The data was extracted with the help of monitors from individual subjects at different stress and emotion intervals. Following that, CNN and Extreme Gradient Boosting were utilized for the classification task. As demonstrated, XGB outperforms CNN with high recall, accuracy, and precision. Visualizing the frequency components and mean values allows one to imagine how the data behaves. Current research has showcased that EEG-based authentication has great potential for a broad range of applications and points to future avenues for investigation, such as multimodal authentication systems or real-time authentication. To make the EEG robust and inclusive, a follow-up study could also include more in-depth feature studies, hyperparameter optimization, or possible demographic effects on the signals. © 2024 IEEE.

关键词： Electroencephalography

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：