检索结果-内蒙古大学图书馆

3rd International Conference on Computer, Artificial Intelligence and Control Engineering, CAICE 2024

作者： Fang, Tianke Hui, Zhenxing Rey, William P Xie, Zhiying Yang, Aihua He, Yuanrong School of Information Technology Mapua University Manila1200 Philippines Big Data Institute of Digital Natural Disaster Monitoring in Fujian Xiamen University of Technology Xiamen Fujian361024 China School of Computing and Information Engineering Xiamen University of Technology Xiamen Fujian361024 China School of Information Media Zhangzhou Vocational College of Science and Technology Zhangzhou Fujian363200 China Hunan Key Laboratory of Remote Sensing Monitoring of Ecological Environment Dongting Lake Area Changsha Hunan410004 China

ISBN: (纸本)9798400716942

In the protection and restoration work of ancient buildings, the detection and analysis of the cracks and defects of the exterior walls of ancient buildings is a crucial task. In this paper, an improved YOLOv7 algorithm is proposed to detect the defects of ancient external walls, aiming to achieve efficient and accurate detection of the defects of ancient external walls. The method involves preprocessing the image using a rectangle clipping algorithm, graying the image using the maximum method, and denoising the image with a median filter. Next, the image is labeled using the LabelImg tool, and the data is split into training, validation, and test sets. During model training, a CBAM-Net module is inserted after several CBS modules in the backbone network, the GIOU loss function is replaced by CIOU, and feature task extraction is performed. The realized attention mechanism is added to the Yolov7 model. The results show that the mAP value of crack detection after the improved algorithm reaches 81.2%. In contrast to the initial YOLOv7 algorithm, the precision of the model has been boosted to 87.2%, exhibiting a 6.0% enhancement in accuracy and an 85.2% improvement in recall rate. The improved YOLOv7 crack detection algorithm reduces network calculation and improves crack classification and recognition ability. This approach is crucial in safeguarding ancient buildings as it enables varying levels of maintenance depending on the severity of cracks. © 2024 ACM.

关键词： Attention mechanism Deep learning Defect detection Image processing YOLOv7

来源：评论

学校读者我要写书评

暂无评论

Dates Fruit Recognition: From Classical Fusion to Deep Learning

引用

Computer Systems science & Engineering 2022年第1期40卷 151-166页

作者： Khaled Marji Alresheedi Suliman Aladhadh Rehan Ullah Khan Ali Mustafa Qamar Department of Computer Science College of ComputerQassim UniversityBuraydahSaudi Arabia Department of Information Technology College of ComputerQassim UniversityBuraydahSaudi Arabia Department of Computing School of Electrical Engineering and Computer Science(SEECS)National University of Sciences and Technology(NUST)IslamabadPakistan

There are over 200 different varieties of dates fruit in the ***,every single type has some very specific features that differ from the *** recent years,sorting,separating,and arranging in automated industries,in fruits businesses,and more specifically in dates businesses have inspired many research *** this regard,this paper focuses on the detection and recognition of dates using computer vision and machine *** experimental setup is based on the classical machine learning approach and the deep learning approach for nine classes of dates *** machine learning includes the Bayesian network,Support Vector Machine,Random Forest,and Multi-Layer Perceptron(MLP),while the Convolutional Neural Network is used for the deep learning *** feature set includes Color Layout features,Fuzzy Color and Texture Histogram,Gabor filtering,and the Pyramid Histogram of the Oriented *** fusion of various features is also extensively explored in this *** MLP achieves the highest detection performance with an F-measure of ***,deep learning shows better accuracy than the classical machine learning *** fact,deep learning got 2%more accurate results as compared to the MLP and the Random *** also show that classical machine learning could give increased classification performance which could get close to that provided by deep learning through the use of optimized tuning and a good feature set.

关键词： features Deep separating

来源：评论

学校读者我要写书评

暂无评论

Defending Against Digital Threats: Machine Learning Techniques for Cyber Persecution Detection 5

Defending Against Digital Threats: Machine Learning Techniqu...

引用

5th International Conference on Artificial Intelligence and Data sciences, AiDAS 2024

作者： Chinnasamy, P. Ayyasamy, Ramesh Kumar Krishnan, Kesavan Kumar, Vijay Jalil, Norazira Binti A Kiran, Ajmeera School of Computing Kalasalingam Academy of Research and Education Department of Computer Science and Engineering Tamilnadu India Faculty of Information and Communication Technology Kampar Malaysia Universiti Malaya Faculty of Computer Science and Information Technology Kuala Lumpur Malaysia Graphic Era Hill University Department of Physics Dehardun India Graphic Era Deemed to be University Dehradun India Mlr Institute of Technology Department of Computer Science and Engineering Hyderabad India

ISBN: (纸本)9798331528553

Cyber persecution has become a widespread problem on the social media. It has resulted in omissions such as suicide and sadness. Content regulation on social media sites is becoming increasingly important. The following work employs natural language processing and Machine Learning techniques to construct a cyber harassment detection model utilizing data from cyberbullying, hate speech tweets from Twitter, and personal assaults from Wikipedia. The optimal approach is investigated by two feature extraction methods and six classifiers (SVM, Random Forest, XGBoost, MLP, and Logistic Regression). This study summarizes the general procedure for detecting cyberbullying and, more crucially, the technique. During experimentation, SVM, Random Forest, and Logistic Regression performed well with an average of 90%. Our model has been created to detect offensive comments on social media websites. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Dimensionality-driven metal to Mott insulator transition in two-dimensional 1T-TaSe2

引用

National science Review 2024年第3期11卷 108-118页

作者： Ning Tian Zhe Huang Bo Gyu Jang Shuaifei Guo Ya-Jun Yan Jingjing Gao Yijun Yu Jinwoong Hwang Cenyao Tang Meixiao Wang Xuan Luo Yu Ping Sun Zhongkai Liu Dong-Lai Feng Xianhui Chen Sung-Kwan Mo Minjae Kim Young-Woo Son Dawei Shen Wei Ruan Yuanbo Zhang State Key Laboratory of Surface Physics New Cornerstone Science Laboratory and Department of PhysicsFudan University Shanghai Qi Zhi Institute Shanghai Research Center for Quantum Sciences Institute for Nanoelectronic Devices and Quantum Computing Fudan University Zhangjiang Fudan International Innovation Center Fudan University State Key Laboratory of Functional Materials for Informatics Shanghai Institute of Microsystem and Information Technology(SIMIT)Chinese Academy of Sciences School of Physical Science and Technology ShanghaiTech University Korea Institute for Advanced Study School of Emerging Technology and Department of Physics University of Science and Technology of China Advanced Light Source Lawrence Berkeley National Laboratory ShanghaiTech Laboratory for Topological Physics Key Laboratory of Materials Physics Institute of Solid State Physics Hefei Institutes of Physical ScienceChinese Academy of Sciences High Magnetic Field Laboratory Hefei Institutes of Physical Science Chinese Academy of Sciences Collaborative Innovation Centre of Advanced Microstructures Nanjing University Department of Physics University of Science and Technology of China and Key Laboratory of Strongly Coupled Quantum Matter Physics Chinese Academy of Sciences National Synchrotron Radiation Laboratory University of Science and Technology of China

Two-dimensional materials represent a major frontier for research into exotic many-body quantum phenomena. In the extreme two-dimensional limit, electron-electron interaction often dominates over other electronic energy scales, leading to strongly correlated effects such as quantum spin liquid and unconventional *** dominance is conventionally attributed to the lack of electron screening in the third dimension. Here, we discover an intriguing metal to Mott insulator transition in 1T-TaSe2that defies conventional wisdom. Specifically, we find that dimensionality crossover, instead of reduced screening, drives the transition in atomically thin 1T-TaSe2. A dispersive band crossing the Fermi level is found to be responsible for the bulk metallicity in the material. Reducing the dimensionality,however, effectively quenches the kinetic energy of these initially itinerant electrons, and drives the material into a Mott insulating *** dimensionality-driven metal to Mott insulator transition resolves the long-standing dichotomy between metallic bulk and insulating surface of 1T-TaSe2. Our work further reveals a new pathway for modulating two-dimensional materials that enables exploring strongly correlated systems across uncharted parameter space.

关键词： two-dimensional materials 1T-TaSe2 dimensionality crossover metal to Mott insulator transition

来源：评论

学校读者我要写书评

暂无评论

Grid Search Hyperparameters Tuning with Supervised Machine Learning for Awngi Language Named Entity Recognition 2

Grid Search Hyperparameters Tuning with Supervised Machine L...

引用

2nd International Conference Computational and Characterization Techniques in Engineering and sciences, IC3TES 2024

作者： Ayalew, Amogne Andualem Bitew, Gizachew Salau, Ayodeji Olalekan Tegegne, Melaku Lake Gupta, Shashi Kant Israr, Mohammad Tegegne, Tesfa Assegie, Tsehay Admassu Mizan-Tepi University School of Computing and Informatics Department of Information Technology Tepi Ethiopia Mizan Tepi University School of Computing Computer Science Department Mizan-Tepi Ethiopia Afe Babalola University Department of Electrical/Electronics and Computer Engineering Ado-Ekiti Nigeria Injibara University College of Engineering and Technology Department of Information Technology Injibara Ethiopia Eudoxia Research University Department of Computer Science and Engineering United States Maryam Abacha American University of Nigeria Hotoro Gra Kano State Nigeria Bahir Dar University Faculty of Computing Bahir Dar Institute of Technology Bahir Dara Ethiopia Injibara University College of Engineering and Technology Department of Computer Science Injibara Ethiopia

ISBN: (纸本)9798350364699

Named entity recognition has emerged as a critical step in recognizing, classifying, and extracting the most significant information from unstructured text without human intervention. It is used in information retrieval, conversation systems, machine translation, data mining, and information extraction. However, the Awngi language lacks a NER system due to inadequate datasets. This study combines feature extraction techniques with supervised machine learning algorithms, including support vector machines, Naïve Bayes, and conditional random fields with a grid search algorithm parameter optimization. These supervised machine-learning models often have various hyperparameters that must be tuned to optimize the model's performance by automatically selecting the parameter. Furthermore, the dataset used for training and testing consists of 20,193 annotated Awngi tokens, collected from Amhara Media Corporation, Awngi Telegram, and Facebook pages. The proposed SVM with grid search algorithm approach provided a promising result in ANER with an accuracy of 92% compared with Naïve Bayes, conditional random field, and decision-tree supervised machine learning models. Finally, the experiment results show that the proposed parameter tuning-based model can detect and categorize Awingi-named entities and outperforms the performance accuracy without parameter adjustment. © 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

JetVision: A Low-Cost Real-Time Depth Estimation System using Jetson computing Platform 9

JetVision: A Low-Cost Real-Time Depth Estimation System usin...

引用

9th IEEE International Conference on Electronics, computing and Communication Technologies, CONECCT 2023

作者： Mandal, Siddhanta Cholachgudda, Kartik E. Chamakura, Lohith V. Biradar, Rajashekhar C. Devanagavi, Geetha D. Reva University School of Computer Science and Engineering Bengaluru India Reva University School of Electronics and Communication Engineering Bengaluru India Reva University School of Computing and Information Technology Bengaluru India

ISBN: (纸本)9798350334395

Stereo vision is an important technique in computer vision that enables depth information to be extracted from a pair of cameras. Depth estimation has numerous applications across various fields. Some of the notable applications include autonomous vehicles, augmented reality (AR) and virtual reality (VR), robotics, 3D reconstruction, surveillance and security and medical imaging. In this research, the authors develop and test a low-cost real-time stereo vision system, JetVision, that uses two Raspberry Pi cameras, a Jetson computing platform and CUDA programming. The algorithm developed for JetVision is designed to efficiently analyze the differences between the images captured by each camera to estimate the depth of objects in the scene using Semi-Global Matching (SGM) method. To achieve real-time performance, the algorithm is implemented using CUDA programming, which enables efficient parallel processing on Jetson's GPU. The algorithm is tested on a dataset of real-world images captured by JetVision and is compared its result with high-grade Intel® RealSenseTM depth camera and ground truth depth information. The results demonstrate that JetVision is capable of accurately estimating depth information in real-time with an FPS ranging between 15 to 28. This research contributes to the field of computer vision by demonstrating the potential of stereo vision algorithms in real-world applications using cost-effective hardware and efficient parallel processing. © 2023 IEEE.

关键词： Stereo vision

来源：评论

学校读者我要写书评

暂无评论

How to Construct Public Timeline for RSA-Formed Time-Lock Cryptography 20th

How to Construct Public Timeline for RSA-Formed Time-Lock ...

引用

20th International Conference on information Security and Cryptology, Inscrypt 2024

作者： Jin, Huixuan Peng, Cong Fu, Jintao Luo, Min Key Laboratory of Aerospace Information Security and Trusted Computing Ministry of Education School of Cyber Science and Engineering Wuhan University Wuhan China Henan Key Laboratory of Network Cryptography Technology Henan Zhengzhou China Inspur Academy of Science and Technology Jinan China

ISBN: (纸本)9789819647330

Time-Locked Puzzle (TLP) and timed-release encryption (T-RE) are both time-based locking mechanisms. The former requires the decryptor to perform a predetermined amount of computation, while the latter requires a trusted server to release the decryption information after a preset time, thereby sending information to the future. As research has progressed, many feature-rich TLPs and TREs have emerged. However, TLP and TRE still face issues in practical deployment. The imbalance in computation load with TLP can result in significant computational burdens in many deployment scenarios, while the timed release in TRE necessitates that the server remains online. To address these issues, our goal is to develop a multifunctional solution to offload the computational burden and allow the server to be offline. Our scheme focuses on RSA-based TLPs and builds upon the sequential TRE schemes proposed by Chvojka and Jager et al.(ESORICS’21). We construct a multi-timeline scheme that aggregates multiple functions and allows the server to be offline. The multi-timeline structure proposed in our scheme, consisting of a main timeline, sub-timelines, and reward timelines, allows different timelines to assume different functions, and the main timeline controls the information release of the remaining timelines to achieve an efficient puzzle-solving release mechanism. The reward timeline provides a transfer account for storing bonuses for entities that solve puzzles, which achieves a self-balancing mechanism for computing resources. Due to the completion of a transaction guaranteed by the adaptor signature and smart contract, which ensure the atomicity of the transaction, the scheme achieves a traceable fair trading mechanism and a public verification mechanism. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： NP-hard

来源：评论

学校读者我要写书评

暂无评论

An Effective Fire Detection Algorithm Based on Improved YOLOv7 24

An Effective Fire Detection Algorithm Based on Improved YOLO...

引用

6th International Conference on Control and Computer Vision, ICCCV 2024

作者： Shen, Jing Wang, Xue Alghazzawi, Daniyal Aldhaheri, Sahar Business School Hubei University Hubei Wuhan430062 China College of Automation and Electronic Engineering Qingdao University of Science and Technology Shandong Qingdao266061 China Faculty of Computing and Information Technology King Abdulaziz University Jeddah21589 Saudi Arabia

ISBN: (纸本)9798400718045

In the critical domain of fire detection, essential for safeguarding public safety and industrial assets, traditional methods have struggled with issues like low accuracy and slow response times, compromising real-time detection capabilities. Our response to this challenge is the development of a novel algorithm built upon the enhanced YOLOv7 framework, which integrates the CBS module with the cutting-edge ConvNeXtBlock to form the CN-B network module—a design that allows for effective longitudinal and transverse normalization, expanding the receptive field to capture a more comprehensive set of global information without the penalty of increased memory usage. This is further bolstered by the introduction of the SIoU loss function, which dramatically improves model convergence speed, ensuring a swift and accurate learning process. Rigorous testing on the BoWFire dataset has proven the algorithm’s superior performance, boasting an average fire detection accuracy of 89.74% and improving detection speed by 2.9 milliseconds per image, outperforming the original YOLOv7 network by a substantial margin and demonstrating a 2.1% increase in precision. Comparative experiments with eight leading networks validate our model’s adherence to the stringent requirements of real-time and accurate fire detection, positioning our improved YOLOv7 network as a significant advancement in the field of fire detection technology. © 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Fires

来源：评论

学校读者我要写书评

暂无评论

Classification of SEMG Signal Based Hand Movements by Using Support Vector Machine 6

Classification of SEMG Signal Based Hand Movements by Using ...

引用

6th IEEE Symposium on Computers and Informatics, ISCI 2024

作者： Zahra, Syeda Rabail Ismail, Sadia Ali, Muhammad Danish Khan, Muhammad Amir Darus, Mohamad Yusof Mazhar, Tahseen Gomal University Institute of Computing & Information Technology Dera Ismail Khan Pakistan University Islamabad Computer Science Comsats Abbottabad Pakistan Universiti Teknologi Mara School of Computing Sciences College of Computing Informatics and Mathematics Selangor Shah Alam40450 Malaysia Virtual University of Pakistan Department of Computer Science Pakistan

ISBN: (纸本)9798350353853

This paper explores the application of Surface Electromyography (SEMG) for hand movement classification using machine learning techniques. SEMG signals play a crucial role in sports medicine, rehabilitation, prosthesis control, and medical diagnosis. The study addresses a multi-class problem, extending the dataset with limited prior exploration. Nine features from SEMG signal's time domain are extracted, including Maximum Amplitude, Root Mean Square, Slope Sign Change, Variance, Simple Square Integral, Zero Crossing, Waveform Length, and Willison Amplitude. Linear Discriminant Analysis is employed for dimensionality reduction as a feature selection technique. The proposed algorithm achieves an overall classification accuracy of 72% for the classification of six different hand movements performed by five healthy subjects using Support Vector Machines. This research contributes to the advancement of SEMG-based hand movement classification in various practical applications. © 2024 IEEE.

关键词： Dimensionality reduction

来源：评论

学校读者我要写书评

暂无评论

Practical fixed-time adaptive fuzzy control of uncertain nonlinear systems with time-varying asymmetric constraints: a unified barrier function based approach

引用

Frontiers of information technology & Electronic Engineering 2024年第9期25卷 1282-1294页

作者： Zixuan HUANG Huanqing WANG Ben NIU Xudong ZHAO Adil M.AHMAD College of Engineering Bohai UniversityJinzhou 121013China College of Mathematical Science Bohai UniversityJinzhou 121013China School of Information Science and Engineering Shandong Normal UniversityJinan 250014China Faculty of Electronic Information and Engineering Dalian University of TechnologyDalian 116024China Communication Systems and Networks Research Group Department of Information TechnologyFaculty of Computing and Information TechnologyKing Abdulaziz UniversityJeddahSaudi Arabia

A practical fixed-time adaptive fuzzy control strategy is investigated for uncertain nonlinear systems with time-varying asymmetric constraints and input quantization. To overcome the difficulties of designing controllers under the state constraints, a unified barrier function approach is employed to construct a coordinate transformation that maps the original constrained system to an equivalent unconstrained one, thus relaxing the time-varying asymmetric constraints upon system states and avoiding the feasibility check condition typically required in the traditional barrier Lyapunov function based control approach. Meanwhile, the “explosion of complexity” problem in the traditional backstepping approach arising from repeatedly derivatives of virtual controllers is solved by using the command filter method. It is verified via the fixed-time Lyapunov stability criterion that the system output can track a desired signal within a small error range in a predetermined time, and that all system states remain in the constraint range. Finally, two simulation examples are offered to demonstrate the effectiveness of the proposed strategy.

关键词： Unified barrier function Time-varying asymmetric state constraints Fuzzy logic systems Fixed-time control Command filter

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：