检索结果-内蒙古大学图书馆

International Conference on Communication and Electronics Systems (ICCES)

作者： Sowmya C S Vibin R Praveen Mannam Lakkakula Mounika Subash Ranjan Kabat Jyoti Prasad Patra Department of ECE Presidency University Bengaluru Karnataka India Department of Electrical and Electronics Engineering CMS College of Engineering and Technology Coimbatore Tamilnadu India Dept of Electrical and Electronics Engineering KKR&KSR Institute of Technology and Sciences Guntur Andhra-Pradesh Lakkakula Mounika School of Computer Science and Artificial Intelligence SR University Warangal Telangana India Department of Electrical Engineering Radhakrishna Institute of Technology and Engineering Bhubaneswar Odisha India OUTR Government of Odisha Mahalaxmi Vihar Ghatikia Techno Campus Faculty Electrical Odisha University of Technology and Research Bhubaneswar Odisha India

Theft of electricity is a major problem that causes financial losses and inconsistent service for paying consumers for power distribution companies all over the world. The safety of the power grid depends on the ability to identify and stop electricity theft. The use of deep learning techniques has shown great promise in recent years, particularly in the areas of computer vision and natural language processing. This study recommends a random forest-based ensemble deep learning method for identifying cases of electricity theft. The proposed ensemble deep learning model leverages the best features of many kinds of deep learning architectures, including stacked Convolutional Neural Networks (CNN)and Long Short-Term Memory (LSTM). Each architecture has its own strengths when it comes to monitoring normal and abnormal electrical use for signs of theft. The final forecast is derived by adding together the predictions of the different models in the random forest ensemble. The ensemble model is trained using a massive dataset of energy usage records and theft information. Information about consumption patterns is extracted using feature engineering methods once the dataset has been preprocessed to get rid of noise and outliers. This preprocessed dataset is used to train the ensemble model, which then optimizes its parameters to reduce prediction errors. We use many measures, including accuracy, precision, recall, and F1-score, to assess the proposed ensemble deep learning model’s performance. Experiments are run against both conventional machine learning methods and standalone deep learning models to prove that the ensemble method is superior. The findings demonstrate that the ensemble model is more accurate and has a greater detection rate, making it suitable for spotting energy theft.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Large Language Models (LLMs) as Traffic Control Systems at Urban Intersections: A New Paradigm

arXiv

引用

arXiv 2024年

作者： Masri, Sari Ashqar, Huthaifa I. Elhenawy, Mohammed Arab American University 13 Zababdeh P.O Box 240 Jenin Palestine Civil Engineering Department Arab American University 13 Zababdeh P.O Box 240 Jenin Palestine Artificial Intelligence Program Fu Foundation School of Engineering and Applied Science Columbia University 500 W 120th St New YorkNY10027 United States CARRS-Q Centre for Data Science Queensland University of Technology 130 Victoria Park Rd Kelvin GroveQLD4059 Australia

This study introduces a novel approach for traffic control systems by using Large Language Models (LLMs) as traffic controllers. The study utilizes their logical reasoning, scene understanding, and decision-making capabilities to optimize throughput and provide feedback based on traffic conditions in real-time. LLMs centralize traditionally disconnected traffic control processes and can integrate traffic data from diverse sources to provide context-aware decisions. LLMs can also delivere tailored outputs using various means such as wireless signals and visuals to drivers, infrastructures, and autonomous vehicles. To evaluate LLMs ability as traffic controllers, this study proposed a four-stage methodology. The methodology includes data creation and environment initialization, prompt engineering, conflict identification, and fine-tuning. We simulated multi-lane four-leg intersection scenarios and generates detailed datasets to enable conflict detection using LLMs and Python simulation as a ground truth. We used chain-of-thought prompts to lead LLMs in understanding the context, detecting conflicts, resolving them using traffic rules, and delivering context-sensitive traffic management solutions. We evaluated the prformance GPT-mini, Gemini, and Llama as traffic controllers. Results showed that the fine-tuned GPT-mini achieved 83% accuracy and an F1-score of 0.84. GPT-mini model exhibited a promising performance in generating actionable traffic management insights, with high ROUGE-L scores across conflict identification of 0.95, decision-making of 0.91, priority assignment of 0.94, and waiting time optimization of 0.92. This methodology confirmed LLMs' benefits as a traffic controller in real-world applications. We demonstrated that LLMs can offer precise recommendations to drivers in real-time including yielding, slowing, or stopping based on vehicle dynamics. This study demonstrates LLMs transformative potential for traffic control, enhancing efficiency and safety a

关键词： Street traffic control

来源：评论

学校读者我要写书评

暂无评论

An effectual IOT coupled EEG analysing model for continuous patient monitoring

引用

Measurement: Sensors 2022年 24卷

作者： Khiani, Simran Mohamed Iqbal, M. Dhakne, Amol Sai Thrinath, B.V. Gayathri, PG. Thiagarajan, R. G H Raisoni College of Engineering and Management Pune India School of Computer Science and Engineering VIT-AP University Andhra Pradesh India Department of Computer Engineering Dr. D. Y. Patil Institute of Engineering Management and Research SPPU Pune Akurdi Pune India Department of EEE Sree Vidyanikethan Engineering College Tirupati India Department of Artificial Intelligence and Data Science Kongunadu College of Engineering and Technology Tamilnadu Trichy India Department of IT Prathyusha Engineering College India

A persistent neurological condition known as epilepsy is characterized by aberrant brain electrical activity. Epilepsy is a disorder characterized by sporadic symptoms and aberrant electrical brain activity that is spasmodic. Such aberrations are supported by non-stationary, non-linear, and multidimensional clinical data. Such data processing and analysis present a variety of computing research difficulties. One of the most often used signals obtained from the human brain in both science and therapeutic settings is the electroencephalogram (EEG). It is an effective resource for offering insightful knowledge of the mechanics of the brain. EEG signal analyses that are precise and thorough are crucial in the diagnosis of brain diseases like epilepsy. The EEG signal is well-liked in the field of biomedical study because of its notable temporal solution, non-invasive recording setup, and convenience with minimal costs. Currently, skilled neurologists make a diagnosis by visually examining EEG information to determine the most likely course of action. It takes time and is error-prone to visually inspect high-dimensional and non-stationary EEG recordings. An automated seizure detector is a crucial tool for better understanding how seizures are generated as well as a relief for medical experts who become exhausted when viewing a continuous, massive, long-term recording. As it records from at least 16 channels, the multichannel automatic seizure detection system is large. It is difficult to use such a method on distantly located subjects. The suggested system gathers data from IoT devices, and patient history-related electronic clinical data that are saved in the cloud were subjected to predictive analytics. The Bi-LSTM-based intelligent healthcare system for tracking and precisely forecasting brain risk. The trade-off between the current system and the current methods is anticipated in the results section. © 2022 The Authors

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

An Efficient Fusion of Infrared and Visible Images using an Optimized Deep Learning Model with Decomposed Non-Subsampled Shearlet Structural Patch Transform

引用

Signal Processing: Image Communication 2025年

作者： P. Murugeswari Sonia Jenifer Rayen M. Ramkumar U. Sakthi Department of Artificial Intelligence and Data Science Karpagam College of Engineering Coimbatore Tamil Nadu 641032 India Sathyabama Institute of Science and Technology Chennai Tamil Nadu 600119 India Electronics and Communication Engineering Department Sri Krishna College of Engineering and Technology Department of Computational Intelligence School of Computing SRM Institute of Science and Technology Chennai Tamil Nadu 603203 India

Multimodal data fusion is pivotal in artificial intelligence, with Infrared and Visible image Fusion (IVF) crucial for enhancing target detection. Thus, an innovative model is proposed for IVF, leveraging optimized deep learning-based feature extraction and a novel fusion model. Initially, a normalized depth wise crested porcupine cross-attention model is introduced to extract complementary details from visible and infrared images. Subsequently, the Decomposed Non-subsampled Shearlet Structural Patch Transform (DNSSPT) model is employed to integrate features and generate the fused image. This involves decomposing the extracted infrared and visible images into different parts such as signal intensity, mean intensity, and signal structure utilizing structural patch decomposition. Next, a membership curve role is employed to precisely determine the weight of average intensity module, thereby minimizing artifacts and maintaining the importance of infrared targets. Additionally, sharpening operations are employed to improve the detail layer of both the visible and infrared images, leading to a fused image with higher contrast. Through subjective and objective experimental evaluations, the developed model surpasses existing fusion techniques by achieving minimum efficiencies of 0.0521,0.0698, 0.0681, 0.0832 seconds with a maximum detection accuracy of 92.71%, 96%, 90.89%, 94%across different datasets such as TNO, MSRS, VIFB, and VOT-RGBT.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Eddy Current Imaging of Titanium Plate Surface Defects Based on Compressed Sensing

Eddy Current Imaging of Titanium Plate Surface Defects Based...

引用

Far East Forum on Nondestructive Evaluation/Testing: New technology & Application (FENDT)

作者： Zhaoyuan Li Bo Ye Xin Xiong Yidan Zhang Weiquan Deng Jun Bao Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China Key Laboratory of Artificial Intelligence of Yunnan Province Kunming China Faculty of Mechanical and Electrical Engineering Kunming University of Science and Technology Kunming China Faculty of Civil Aviation and Aeronautics Kunming University of Science and Technology Kunming China

ISBN: (纸本)9781665401722

Titanium has excellent comprehensive properties, and its alloys are widely used in aerospace, petrochemical, bioengineering and other industries, among which plate is the most widely used. Before using titanium plate, visual eddy current testing of surface defects is particularly important. However, a large amount of data will be generated in the detection process, most of which are redundant data unrelated to defect imaging. Imaging with this data is not only inefficient, but also a waste of detection resources. In order to solve this problem, this paper proposes an eddy current imaging method of titanium plate surface defects based on compressed sensing. The induced voltage signal in the process of eddy current probe C-scan is sampled at a frequency far lower than that required by Nyquist sampling theorem, which shortens the sampling time and improves the sampling efficiency. The sampled signal is reconstructed by reconstruction algorithm, which can achieve almost the same effect as the original signal. The experimental results show that the proposed method only uses 3 / 8 of the original signal for imaging, and the reconstruction algorithm used in this method has faster reconstruction speed, lower reconstruction error and better reconstruction effect than other mainstream reconstruction algorithms. The use of this method improves the efficiency of the eddy current imaging detection of titanium plate surface defects, and compresses the space occupied by the imaging data. It can not only provide important technical support for optimizing the eddy current detection process of titanium plate surface defects, but also has important significance for the eddy current detection of other metals.

关键词： Surface reconstruction Image edge detection Imaging Titanium Reconstruction algorithms Eddy currents Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Heterophily-Aware Graph Attention Network

arXiv

引用

arXiv 2023年

作者： Wang, Junfu Guo, Yuanfang Yang, Liang Wang, Yunhong State Key Laboratory of Software Development Environment Beihang University Beijing100191 China School of Computer Science and Engineering Beihang University Beijing100191 China Shen Yuan Honors College Beihang University Beijing100191 China School of Artificial Intelligence Hebei University of Technology Tianjin300401 China Hebei Province Key Laboratory of Big Data Calculation Hebei University of Technology Tianjin300401 China

Graph Neural Networks (GNNs) have shown remarkable success in graph representation learning. Unfortunately, current weight assignment schemes in standard GNNs, such as the calculation based on node degrees or pair-wise representations, can hardly be effective in processing the networks with heterophily, in which the connected nodes usually possess different labels or features. Existing heterophilic GNNs tend to ignore the modeling of heterophily of each edge, which is also a vital part in tackling the heterophily problem. In this paper, we first propose a heterophily-aware attention scheme and reveal the benefits of modeling the edge heterophily, i.e., if a GNN assigns different weights to edges according to different heterophilic types, it can learn effective local attention patterns, enabling nodes to acquire appropriate information from distinct neighbors. Then, we propose a novel Heterophily-Aware Graph Attention Network (HA-GAT) by fully exploring and utilizing the local distribution as the underlying heterophily, to handle the networks with different homophily ratios. To demonstrate the effectiveness of the proposed HA-GAT, we analyze the proposed heterophily-aware attention scheme and local distribution exploration, by seeking an interpretation from their mechanism. Extensive results demonstrate that our HA-GAT achieves state-of-the-art performances on eight datasets with different homophily ratios in both the supervised and semi-supervised node classification tasks. Copyright © 2023, The Authors. All rights reserved.

关键词： Graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based detection of incisal translucency patterns

引用

Journal of Prosthetic Dentistry 2025年

作者： Shetty, Sthithika Gali, Sivaranjani R, Venkatesh MS, Jeswin MN, Thippeswamy Postgraduate student Department of Prosthodontics and Crown & Bridge Faculty of Dental Sciences M.S. Ramaiah University of Applied Sciences Bangalore India Head Department of Prosthodontics and Crown & Bridge Faculty of Dental Sciences M.S. Ramaiah University of Applied Sciences Bangalore India Graduate student Department of Computer Science & Engineering (Artificial Intelligence & Machine Learning) Ramaiah Institute of Technology Bangalore India Principal Dr. Ambedkar Institute of Technology Bangalore India

Statement of problem: The evaluation of incisal translucency in anterior teeth greatly influences esthetic treatment outcomes. This evaluation is mostly subjective and often overlooked among dental professionals. The application of artificial intelligence-based models to detect the incisal translucency of anterior teeth may be of value to dentists in their restorative dental practice, but studies are lacking. Purpose: The purpose of this study was to assess the accuracy of deep learning models in predicting the translucency patterns of anterior teeth. Material and methods: Approximately 240 Joint Photographic Experts Group (JPEG) images of anterior teeth from participants over 18 years were collected using a smartphone. These images were resized to 224×224 pixels and classified by the presence or absence of translucency. Augmentation techniques enhanced the training dataset, and a 3-model deep learning approach was used: YOLOv5 detected central incisors, Vision Transformers (ViT) identified translucency, and U-Net segmented the translucent areas. The images were split 80 to 20 for training and testing, with performance evaluated using accuracy, precision, recall, F1 score, confusion matrix, and dice scores. Results: YOLOv5 achieved a precision of 1.00 at a confidence threshold of 0.910. The ViT system showed an accuracy of 91.66%, with 58 of 64 images predicting correctly with an F1 score of 94.83%. U-Net segmentation after training with annotated images achieved an accuracy of 91% with a dice score of 0.948. Conclusions: The integration of YOLOv5 for detection, ViT for classification, and U-Net for segmentation demonstrates a comprehensive approach to addressing the classification of incisal translucencies. By leveraging the strengths of deep learning models, high accuracy and precision can be achieved in detecting the incisal translucency patterns of anterior teeth. © 2025 Editorial Council for The Journal of Prosthetic Dentistry

关键词：

来源：评论

学校读者我要写书评

暂无评论

Carbon fiber reinforced polymer defect detection using magnetic induction tomography method

Carbon fiber reinforced polymer defect detection using magne...

引用

Far East Forum on Nondestructive Evaluation/Testing: New technology & Application (FENDT)

作者： Honggui Cao Bo Ye Siqi Luo Jun Bao Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming China Yunnan Provincial Key Laboratory of Artificial Intelligence Kunming China Faculty of Civil Aviation and Aeronautics Kunming University of Science and Technology Kunming China

ISBN: (纸本)9781665401722

Carbon Fiber Reinforced Polymer (CFRP) is widely used in aerospace, military and other fields due to its excellent performance. In order to meet the requirements of rapid detection, location and visualization of defects in carbon fiber composite reinforced polymer. The defect detection method of carbon fiber reinforced polymer based on magnetic induction tomography (MIT) was investigated. MIT of four-layer unidirectional carbon fiber reinforced polymer laminates model was established by COMSOL Multiphysics, a planar electromagnetic sensor array for CFRP laminates was designed, the sensitivity matrix was calculated based on the reciprocity theorem and correction factor P was introduced to eliminate the influence of negative sensitivity on image reconstruction, then Tikhonov regularization algorithm and Landweber algorithm were used for image reconstruction of CFRP laminates defects. Thus, the results show that the designed planar sensor array can effectively detect the defects of CFRP laminates, at the same time, image reconstruction based on improved sensitivity has better effect on defect detection, and the shape and location of defects are more accurate; It shows that MIT is feasible for defect detection of carbon fiber reinforced polymer.

关键词： Optical fiber sensors Sensitivity Transmission line matrix methods Magnetic resonance imaging Tomography Polymers Laminates

来源：评论

学校读者我要写书评

暂无评论

Tcgfusion: A Network for Pet-Mri Fusion Based on Gan and Transformer

SSRN

引用

SSRN 2022年

作者： Fan, Chao Chen, Zhixiang Lin, Hao Wang, Xiao School of Artificial Intelligence and Big Data Henan University of Technology Henan Province Zhengzhou City China School of Information Science and Engineering Henan University of Technology Henan Province Zhengzhou City China Key Laboratory of Grain Information Processing and Control Ministry of Education Henan Province Zhengzhou City China

We provide an end-to-end unsupervised training model called TCGFusion: a network for PET-MRI fusion based on GAN and transformer. We set the generator to a dual-path layout to assure the gathering of both global and detailed image information. To guarantee the coarse scale picture extraction and prevent the high computational cost, the global path T-Path utilizes Grid-former (a residual grid transformer structure). The detailed path C-Path is DenseNet, and its multi-cascade structure captures image details well. Additionally, we set up a dual adversarial mechanism between the fused image and the source image, creating a dual discriminator to determine the distribution of structural similarity between the fused result and the source image, and using patch-SSIM to reinforce the semantic constraints between them and promote the image details. According to our extensive studies on public datasets, medical image fusion demonstrate the superiority of our TCGFusion compared to the state-of-the-art image fusion algorithms. © 2022, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Dfe-Net: Dual-Branch Feature Extraction Network for Enhanced Segmentation in Skin Lesion

SSRN

引用

SSRN 2022年

作者： Fan, Chao Yang, Litao Lin, Hao Qiu, Yingying School of Artificial Intelligence and Big Data Henan University of Technology Henan Province Zhengzhou City China Key Laboratory of Grain Information Processing and Control Ministry of Education Henan Province Zhengzhou City China School of Information Science and Engineering Henan University of Technology Henan Province Zhengzhou City China

Skin lesion segmentation is a critical method for extracting pathological information from dermoscopy images, which is of great significance for lesion location, recognition, monitoring and treatment. Due to the varied sizes of lesions, different forms and colors, and blurring boundaries, the present detection methods are unable to accurately anticipate local features, which has an impact on the segmentation accuracy. Therefore, the Dual-branch Feature Extraction Network (DFE-Net) is suggested in this paper. We designed two types of encoders based on the Transformer and Efficient Channel Attention (ECA) module, one for extracting the global feature and another for the local feature, before fusing and decoding. Simultaneously, the skip connections from the Enhanced ECA Feature Extraction Modules to decoders are designed to reduce feature loss during decoding, this will restore the boundaries and local features to the greatest extent possible. We test the model on three publicly available skin lesion datasets: PH2, ISIC-2016, and ISIC-2018. The results show that our model is superior to other previous methods, and the segmentation characteristics are more accurate representations of the actual lesion. © 2022, The Authors. All rights reserved.

关键词： Feature extraction

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：