检索结果-内蒙古大学图书馆

4th international conference for Emerging Technology, INCET 2023

作者： Joshi, Devesh Mohd, Noor University Department of Computer Science and Engineering Uttarakhand Dehradun India

ISBN: (纸本)9798350335750

the Automatic Number Plate recognition (ANPR) systems typically consist of a camera, image processing software, and a database system. When a car passes by, the camera snaps a picture of its license plate. the image processing software then uses OCR technology to retrieve the license plate number from the picture. the extracted plate number is then compared against a database of registered vehicles, and any relevant information associated with the plate number is retrieved. this information can include the vehicle's make and model, owner information, and any applicable fines or penalties. By automatically capturing license plate numbers, ANPR can help identify stolen vehicles, track the movements of known criminals, and enforce traffic violations such as speeding and running red lights. In addition, ANPR can also be used for parking enforcement, enabling authorities to enforce regulations. ANPR is also widely used in the toll collection industry, where drivers can be automatically charged for using toll roads without the need for physical barriers or manual inspection of license plates. Overall, ANPR technology is a valuable tool for improving road safety, reducing crime, and increasing efficiency in a variety of applications. However, like all technologies, ANPR has the potential to be misused, and its use must be balanced against privacy concerns and the right to protect personal information. © 2023 IEEE.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

One-Staged Attention-Based Neoplasms recognition Method for Single-Channel Monochrome computer Tomography Snapshots

引用

pattern recognition AND image ANALYSIS 2022年第3期32卷 645-650页

作者： Samarin, A. Savelev, A. Toropov, A. Dzestelova, A. Malykh, V Mikhailova, E. Motyko, A. ITMO Univ St Petersburg 197101 Russia St Petersburg Electrotech Univ LETI St Petersburg 197022 Russia

computer tomography is most commonly used for diagnosing lung cancer, which is one of the deadliest cancers in the world. Online services that allow users to share their single-channel monochrome images, in particular computer tomography scans, in order to receive independent medical advice are becoming wide-spread these days. In this paper, we propose an optimization for the previously known two-staged architecture for detecting cancerous tumors in computer tomography scans that demonstrates the state-of-the-art results on Open Joint Monochrome Lungs computer Tomography dataset. Modernized architecture allows to reduce the number of weights of the neural network based model (4 920 073 parameters vs. 26 468 315 in the original model) and its inference time (0.38 vs. 2.15 s in the original model) without loss of neoplasms recognition quality (0.996 F-1 score). the proposed results were obtained using heavyweight encoder elimination, special combined loss function and watershed based method for the automated dataset markup and a consistency regularization approach adaptation that are described in the current paper.

关键词： biomedical images processing neoplasms recognition computer tomography snapshots classification

来源：评论

学校读者我要写书评

暂无评论

Design and Optimization of a Dynamic Train image Acquisition System Utilizing Line Scan Camera Technology 4

Design and Optimization of a Dynamic Train Image Acquisition...

引用

4th international conference on Advances in Electrical Engineering and computer Applications, AEECA 2023

作者： Li, Changgeng Xu, Feng Li, Wei Sun, Xiaodong Yuan, Bo Crrc Qingdao Sifang Co. Ltd Qingdao266111 China

ISBN: (纸本)9798350308082

In response to the demand for efficient and high-quality daily maintenance and inspection of modern railway trains, the utilization of AI-based machine vision technology has been proven to significantly enhance maintenance efficiency and quality. image quality plays a critical role in visual inspection. thus, this study presents the design of a dynamic train exterior image acquisition system based on line scan camera technology. through a comprehensive analysis of maintenance requirements and business scenarios, the hardware structure and functional framework of the system are meticulously developed. the image acquisition module is strategically deployed in a 360° arrangement to enable panoramic image acquisition of the train's exterior. Leveraging the imaging principle of the line scan camera, two sets of magnetic cylinders are employed to measure the train's velocity multiple times, ensuring continuous correction of the image acquisition line frequency for stable image capture. Furthermore, a character recognition algorithm for train numbers is extensively researched, enabling accurate identification of the train number through image acquisition, preprocessing, and OCR recognition algorithms. the proposed system integrates professional image acquisition and processing techniques, enabling high-resolution and high-precision stable acquisition of dynamic train images. It provides crucial data support for precise detection of train exterior conditions and offers reliable technical means for train trajectory monitoring and safety management. © 2023 IEEE.

关键词： image acquisition

来源：评论

学校读者我要写书评

暂无评论

Real-Time Automated Pothole Detection and Localization with Deep Learning and Geolocation Integration for Improved Road Safety and Maintenance 4th

Real-Time Automated Pothole Detection and Localization with ...

引用

4th international conference on Advanced Network Technologies and Intelligent Computing-ANTIC-Annual

作者： Rao, Abhishek S. Karthik, Pai B. H. Shyvalanai, B. S. Nayak, Trupthi D. Vaishnavi, R. thanisha NITTE NMAM Inst Technol NMAMIT Dept Informat Sci & Engn Nitte 574110 Karnataka India

ISBN: (纸本)9783031837920;9783031837937

Real-time identification of road damage, particularly potholes, is essential for enhancing road safety and minimizing vehicle damage. Traditional road inspection methods are often labor-intensive and inefficient. To overcome these limitations, we propose an automated system that leverages deep learning to detect potholes in real-time from video input. this system can detect potholes under varying lighting conditions, specifically the YOLO (You Only Look Once) model, utilizing advanced computer vision techniques, including daylight, darkness, and vehicle headlights. the system aims to improve road monitoring by automatically identifying potholes and notifying authorities of their locations. By integrating geolocation services, the system pinpoints pothole locations using latitude and longitude coordinates and sends real-time alerts via a Telegram bot. Additionally, image enhancement techniques are employed to optimize detection performance in low-light conditions. the system has demonstrated high accuracy in detecting large potholes and identifying multiple potholes within a single frame. through automated pothole detection and location sharing, this solution has the potential to significantly enhance road maintenance efficiency, thereby improving road safety and reducing accidents caused by road damage.

关键词： Pothole Detection Real-time Video processing Deep Learning computer vision Low-light image Enhancement

来源：评论

学校读者我要写书评

暂无评论

image Segmentation Using Attribute Network 8

Image Segmentation Using Attribute Network

引用

8th international conference on Signal and image processing, ICSIP 2023

作者： Zheng, Tianrong Dai, Fang Wang, Xiaoxia Xi'An University of Technology School of Science Xi'an China

ISBN: (纸本)9798350397932

image segmentation is one of the key research topics in image analysis and pattern recognition. In recent years, with the rapid development of complex network theory, many researchers have applied complex network theory to the fields of image analysis and computer vision. Existing methods for image segmentation using complex networks only consider the connections between nodes, ignoring the characteristics of the nodes, which means that some important information of the image is lost during the modeling process, and affects the image segmentation performance. In this paper, both the connections between nodes and the characteristics of the nodes themselves are considered, an image topology network is constructed for the connections between nodes, and an attribute network is constructed for the features of nodes. thus, an image segmentation algorithm using attribute networks (SAINS) is proposed in this paper. the SAINS method can effectively segment image without specifying the number of segmented regions. Experimental results show that the proposed algorithm SAINS has better image segmentation accuracy than GFCNIS and LDCS algorithm. © 2023 IEEE.

关键词： image segmentation

来源：评论

学校读者我要写书评

暂无评论

KANICE: Kolmogorov-Arnold Networks with Interactive Convolutional Elements 24

KANICE: Kolmogorov-Arnold Networks with Interactive Convolut...

引用

4th international conference on AI-ML Systems, AIMLSystems 2024

作者： Ferdaus, Md Meftahul Abdelguerfi, Mahdi Ioup, Elias Dobson, David Niles, Kendall N. Pathak, Ken Sloan, Steven University of New Orleans New Orleans United States Center for Geospatial Sciences Naval Research Laboratory Stennis Space Center Mississippi United States US Army Corps of Engineers Engineer Research and Development Center Vicksburg United States

ISBN: (纸本)9798400711619

We introduce KANICE (Kolmogorov-Arnold Networks with Interactive Convolutional Elements), a novel neural architecture that combines Convolutional Neural Networks (CNNs) with Kolmogorov-Arnold Network (KAN) principles. KANICE integrates Interactive Convolutional Blocks (ICBs) and KAN linear layers into a CNN framework. this leverages KANs’ universal approximation capabilities and ICBs’ adaptive feature learning. KANICE captures complex, non-linear data relationships while enabling dynamic, context-dependent feature extraction based on the Kolmogorov-Arnold representation theorem. We evaluated KANICE on four datasets: MNIST, Fashion-MNIST, EMNIST, and SVHN, comparing it against standard CNNs, CNN-KAN hybrids, and ICB variants. KANICE consistently outperformed baseline models, achieving 99.35% accuracy on MNIST and 90.05% on the SVHN dataset. Furthermore, we introduce KANICE-mini, a compact variant designed for efficiency. A comprehensive ablation study demonstrates that KANICE-mini achieves comparable performance to KANICE with significantly fewer parameters. KANICE-mini reached 90.00% accuracy on SVHN with 2,337,828 parameters, compared to KANICE’s 25,432,000. this study highlights the potential of KAN-based architectures in balancing performance and computational efficiency in image classification tasks. Our work contributes to research in adaptive neural networks, integrates mathematical theorems into deep learning architectures, and explores the trade-offs between model complexity and performance, advancing computer vision and pattern recognition. the source code for this paper is publicly accessible through our Github repository (https://***/mferdaus/kanice). © 2024 Copyright held by the owner/author(s).

关键词： image classification

来源：评论

学校读者我要写书评

暂无评论

Defect Detection in Metal Surfaces Using computer vision 1

引用

4th international conference on Recent Trends in Machine Learning, IoT, Smart Cities, and Applications, ICMISC 2023

作者： Singh, Krishna Kumar Ghosh, Manish Symbiosis Centre for Information Technology Pune India

ISBN: (数字)9789819994427

ISBN: (纸本)9789819994410

Encoder models have shown remarkable success in various computer vision operations like object detection, image classification, and semantic segmentation. However, the results of one model showed that it was underfitting as it performed better on the validation set than on the training set. To gain insights into how the model was processing input images at different layers and which image features were being detected by different filters in the convolutional layers, an algorithm was used to visualize the activation maps for the output of convolutional layers in the model. To address the underfitting issue, the previous model was converted into a single feature-vector using global average pooling and used as a single layer in another deep learning model. As a result, the new model displayed a batch of images along with their corresponding segmentation masks and predicted segmentation masks generated by the encoder model. the output was a 3 × 1 image grid that included the original image, its ground truth segmentation mask, and the predicted segmentation mask. To optimize the model further, an algorithm was applied that trained an image encoder model on training data and evaluated it on validation data for a number of epochs. the algorithm then displayed a batch of samples with their ground truth and predicted segmentation masks. the results for this model were better than the previous model, as it was performing better with the train set as compared to the validation set. Overall, the usage of an encoder model and subsequent modifications allowed for improved performance in image segmentation tasks. © the Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Ulcerative Colitis image Classification Using Federated Deep Learning 4th

Ulcerative Colitis Image Classification Using Federated Deep...

引用

4th international conference on Intelligent Systems and pattern recognition-ISPR

作者： Al-Refai, Mohammed Alkhaza'leh, Shahed Alzu'bi, Ahmad Jordan Univ Sci & Technol Fac Comp & Informat Technol Dept Comp Sci Irbid 22110 Jordan

ISBN: (纸本)9783031821523;9783031821530

Ulcerative Colitis is a chronic inflammatory bowel disease characterized by varying degrees of disease severity, often measured using the Mayo endoscopic score. the handling of medical data, encompassing sensitive personal health records and clinical information, necessitates stringent privacy protections. this study introduces a federated deep learning framework that leverages fine-tuned deep learning models trained via transfer learning on a dataset of endoscopic images. Employing DenseNet-121 as the foundational architecture, this approach enables the extraction and encoding of generic descriptors from ulcerative colitis images across multiple clients. We evaluated the training efficiency of our federated learning model in comparison to traditional centralized model, where the federated learning model outperformed the centralized approach by achieving an F1-score of 78% and a Quadratic Weighted Kappa score of 71%, and enhancing convergence speeds in terms of the reduction in the number of training rounds required. the results affirm the model's capability in accurately diagnosing ulcerative colitis while ensuring the confidentiality of patient data, underscoring the viability of federated learning in sensitive healthcare applications.

关键词： Federated Learning Ulcerative Colitis image Classification Neural Networks

来源：评论

学校读者我要写书评

暂无评论

A Remote Sensing image Defogging Method Based on Improved CycleGAN Network 4

A Remote Sensing Image Defogging Method Based on Improved Cy...

引用

4th international conference on computer vision, image and Deep Learning, CVIDL 2023

作者： Zhong, Mianqing Wang, Xue Wang, Jun Kang, Junmei Xi'an China College of Resources Environment and History Culture Xianyang Normal University Xianyang China Second Monitoring and Application Center China Earthquake Administration Xi'an China

ISBN: (纸本)9798350326444

this article presents a methodology for dehazing remote sensing images using deep learning and multi-temporal remote sensing data. the CycleGAN network has been selected as the foundation. We created a sample set of satellite images that are roughly aligned, depicting both smoggy and non-smoggy conditions, using multi-temporal remote sensing resources. In the proposed defogging algorithm, the perceptual loss of cyclic consistency is added, and the trained VGG16 network is utilized to input the original and generated images simultaneously. this allows for the features of a certain layer to be obtained and the loss to be calculated. the structure is beneficial for minimizing the reliance on paired training data, preserving image details, and enhancing clarity. the experimental results on GF1, GF2, and GF7 satellite images demonstrate that the proposed method produces high-quality, clear images. the system is designed to process remote sensing images without requiring extensive processing of complex data and tags. © 2023 IEEE.

关键词： image enhancement

来源：评论

学校读者我要写书评

暂无评论

Glaucoma Scanner: Harnessing Deep CNN for Early Glaucoma Detection 4

Glaucoma Scanner: Harnessing Deep CNN for Early Glaucoma Det...

引用

4th IEEE international conference on Electrical, computer, and Energy Technologies, ICECET 2024

作者： Islam, Aabrar Haque, Ayen Aziza Tasnim, Najifa Anantee, Simin Waliza Karim, Dewan Ziaul Brac University Computer Science & Engineering Dhaka Bangladesh

ISBN: (纸本)9798350395914

Glaucoma is a severe eye condition that can lead to progressive vision impairment if left untreated. Diagnosis and monitoring of glaucoma at an initial stage is critical for effective treatment. However, the diagnosis is complex with bare eyes which requires multiple checkups and tests. image processing is important for diagnosing glaucoma because it provides valuable information about the intricate structure of the eye which helps improve the accuracy of diagnosis and allows for earlier detection of the disease. this portrays the requirement for further research in this field. We aim to explore various image-processing models used for image classification and develop an efficient model that can be used in the detection of glaucoma. In this paper, we have proposed a Custom CNN model with 22 layers based on deep learning for glaucoma diagnosis, which detects from the fundus images whether the person has glaucoma or not. the model has been trained using a dataset containing 4000 fundus images, each with two categories, which are Glaucoma and Non-Glaucoma. the dataset has been used on the Custom CNN model and six other pre-trained models. Our proposed model has been able to successfully classify the images with an accuracy of 98.71 % which was the highest among all the models despite having fewer parameters than other models. © 2024 IEEE.

关键词： Convolutional neural networks

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：