检索结果-内蒙古大学图书馆

27th International Symposium on Design & Diagnostics of Electronic Circuits & systems (DDECS)

作者： Turco, v. Ruospo, A. Sanchez, E. Reorda, M. Sonza Politecn Torino DAUIN Turin Italy

ISBN: (纸本)9798350359343

Computational models based on deep learning are today integrated in many safety-critical domains. These algorithms, such as deep neural networks (DNNs), are rapidly growing in size, reaching billions or even trillions of parameters. This factor brings big challenges not only for performance goals but also for dependability aspects such as reliability. The larger the model, the more challenging the reliability assessment becomes. It is now crucial to develop new test approaches supported by acceptable computational costs for the detection of random-hardware faults such as permanent faults, which may change the predictions of DNNs. The aim of this paper is to leverage tensor-related metrics to early detect faulty behaviors during the inference of DNNs. This involves calculating metrics applied to tensors across various domains (such as image processing, audio analysis, and regression) on the Output Feature Maps (OFMs) of a layer. This analysis allows knowing in advance the effect that a permanent fault will have on the output of the DNN application. The effectiveness of the approach has been experimentally demonstrated by means of software fault injection campaigns considering faults affecting weights of Convolutional Neural Networks (CNNs), i.e., ResNet20 and MobileNetv2. The quality of the metrics is discussed in terms of the trade-off between energy consumption and the ability to differentiate between critical and non-critical faults.

关键词： Deep Neural Networks Reliability Fault Injection Functional Safety

来源：评论

学校读者我要写书评

暂无评论

Robotic Manipulator Software and Hardware Development Using Recognition algorithms

Robotic Manipulator Software and Hardware Development Using ...

引用

2024 Conference of Young Researchers in Electrical and Electronic Engineering, ElCon 2024

作者： Borovkov, Grigoriy S. Martirosyan, Karina v. Protasov, Yakov A. Saint Petersburg Mining University Department of 'System Analysis and Control' Saint Petersburg Russia North-Caucasus Federal University Branch in Pyatigorsk Control Systems and Information Technologies Department Pyatigorsk Russia

ISBN: (纸本)9798350360646

The enterprises digitalization determines the development trends of the industrial sector. There is a need to reduce the percent of human participation in processes associated with conveyor production, which requires the high-tech solutions integration. Due to the lack of quality control on the assembly line, enterprises produce a large percentage of defective products. The article discusses a possible solution to the problem of image processing for monitoring defects in products moving along a conveyor belt. The use of neural network technologies allows us to identify defects in production in real time, and the use of a robotic arm allows us to immediately remove such products from the assembly line. As a result of the research, it is planned to implement a software and hardware complex of a robotic manipulator using computer vision technologies that can distinguish objects and mechanically influence them. © 2024 IEEE.

关键词： Arduino automated sorting system computer vision data collection and analysis defect detection image classification neural networks product quality control robotics in manufacturing statistics

来源：评论

学校读者我要写书评

暂无评论

An Analysis of image Features Extracted by CNNs to Design Classification Models for COvID-19 and Non-COvID-19

引用

JOURNAL OF SIGNAL processing systems FOR SIGNAL image AND vIDEO TECHNOLOGY 2023年第2-3期95卷 101-113页

作者： Teodoro, Arthur A. M. Silva, Douglas H. Saadi, Muhammad Okey, Ogobuchi D. Rosa, Renata L. Al Otaibi, Sattam Rodriguez, Demostenes Z. Univ Fed Lavras Dept Comp Sci Lavras MG Brazil Univ Cent Punjab Dept Elect Engn Lahore 54000 Pakistan Univ Fed Lavras Dept Syst Engn & Automat Lavras MG Brazil Taif Univ Coll Engn Dept Elect Engn At Taif 21944 Saudi Arabia

The SARS-Cov-2 virus causes a respiratory disease in humans, known as COvID-19. The confirmatory diagnostic of this disease occurs through the real-time reverse transcription and polymerase chain reaction test (RT-qPCR). However, the period of obtaining the results limits the application of the mass test. Thus, chest X-ray computed tomography (CT) images are analyzed to help diagnose the disease. However, during an outbreak of a disease that causes respiratory problems, radiologists may be overwhelmed with analyzing medical images. In the literature, some studies used feature extraction techniques based on CNNs, with classification models to identify COvID-19 and non-COvID-19. This work compare the performance of applying pre-trained CNNs in conjunction with classification methods based on machine learning algorithms. The main objective is to analyze the impact of the features extracted by CNNs, in the construction of models to classify COvID-19 and non-COvID-19. A SARS-Cov-2 CT data-set is used in experimental tests. The CNNs implemented are visual geometry group (vGG-16 and vGG-19), inception v3 (Iv3), and EfficientNet-B0 (EB0). The classification methods were k-nearest neighbor (KNN), support vector machine (SvM), and explainable deep neural networks (xDNN). In the experiments, the best results were obtained by the EfficientNet model used to extract data and the SvM with an RBF kernel. This approach achieved an average performance of 0.9856 in the precision macro, 0.9853 in the sensitivity macro, 0.9853 in the specificity macro, and 0.9853 in the F1 score macro.

关键词： COvID-19 non-COvID-19 Computed tomography images Transfer learning Convolutional neural networks Machine learning

来源：评论

学校读者我要写书评

暂无评论

Dog Breed Detection using YOLOv7 and WOA-CNN 10

Dog Breed Detection using YOLOv7 and WOA-CNN

引用

10th International Conference on Communication and Signal processing, ICCSP 2024

作者： Singh, Aniket Kothari, Rushil Arulalan, v. Srm Institute of Science and Technology Department of Computing Technologies Chengalpattu District Tamil Nadu Kattankulathur603203 India

ISBN: (纸本)9798350353068

The need to accurately identify dog breeds is important due to their popularity and role as companions. With an increase in the variety of dog breeds worldwide, there is a pressing demand for better classification methods. Advances in artificial intelligence have resulted in the creation of a new model that combines YOLOv7 with the Whale Optimization Algorithm (WOA) in a Convolutional Neural Network (CNN) setup, designed specifically for detecting dog breeds. This model uses the extensive Stanford Dogs dataset, which includes 120 unique breeds, and has been thoroughly tested. It has shown to be more accurate and precise than current methods. The effectiveness of this model in identifying dog breeds marks a significant improvement in the field and suggests its usefulness in various areas such as veterinary services, dog behavior studies, and improved pet management systems. Additionally, this method showcases the benefits of merging advanced machine learning techniques with nature-inspired algorithms to address complex identification challenges, making a significant contribution to the development of AI-based solutions in animal identification and related areas. © 2024 IEEE.

关键词： image recognition

来源：评论

学校读者我要写书评

暂无评论

Deep Learning-Driven Surveillance for Intrusion Identification using Proactive Intelligent video Recognition 20

Deep Learning-Driven Surveillance for Intrusion Identificati...

引用

20th IEEE International Conference on Advanced video and Signal-Based Surveillance (AvSS)

作者： Challagundla, Yagnesh Parvatham, Shreeya Dheera Mohanty, Sachi Nandan Ravindra, J. v. R. VIT AP Univ Sch Comp Sci & Engn SCOPE Amaravati India Vardhaman Coll Engn Dept Elect & Commun Engn Hyderabad Telangana India

ISBN: (纸本)9798350374292;9798350374285

In today's security-conscious environment, the need for effective real-time weapon detection systems is paramount, especially in public spaces and sensitive areas. The primary objective of this research is to create a real-time system for detecting weapons utilizing advanced deep learning techniques, namely vGG16 and Faster RCNN. The system's objective is to precisely detect and categorize weapons from photos and video data, offering prompt notifications and improving security procedures. The project's backdrop explores the difficulties encountered, including data gathering, precision standards, immediate processing, confidentiality, and deployment considerations. The system achieves excellent accuracy and real-time detection capabilities by building a comprehensive dataset manually and training the models using GPUs and modern technologies. Using Python as the programming language provides flexibility and simplicity in development, making use of Python's modules such as OpenCv for image processing and Keras for deep learning models. The Tkinter framework enables the creation of a graphical user interface (GUI) that supports various user operations, such as uploading datasets, generating models, processing images and videos, detecting weapons, and visualizing results. The methodology employs a systematic approach, encompassing stages such as data preprocessing, model building, training, testing, and result analysis. The combination of vGG16 and Faster RCNN algorithms demonstrates a compromise between speed and accuracy, with Faster RCNN exhibiting greater performance in real-time detection. The project aims to achieve several objectives, including the compilation of a dataset, training a model, evaluating accuracy, implementing real-time detection, and establishing alerting methods. The system's range encompasses a wide range of surroundings, lighting situations, and orientations, allowing it to be flexible and suitable for a variety of security applications. To sum

关键词： Real-time weapon detection Deep learning algorithms vGG16 Faster RCNN Speed vs accuracy tradeoff Real-time detection Performance analysis

来源：评论

学校读者我要写书评

暂无评论

Pothole Detection Approach Based on Deep Learning algorithms 5th

Pothole Detection Approach Based on Deep Learning Algorithms

引用

5th International Conference on Soft Computing and Signal processing, ICSCSP 2022

作者： Chowdary, Y. Aneesh Teja, v. Sai Krishna, v. vamsi Naidu, N. venkaiah Karthika, R. Department of Electronics and Communication Engineering Amrita School of Engineering Amrita Vishwa Vidyapeetham Coimbatore India

ISBN: (纸本)9789811986680

In advancing nations like India, fixing potholes on streets may be a critical responsibility, as accidents can occur owing to their presence. In this way, it is straightforward to distinguish them to guarantee the security of individuals. Many work inquiries have been conducted to date in order to efficiently identify the potholes, but the majority are economically not feasible. In current history, using deep learning algorithms to detect potholes became more popular. Convolutional neural network (CNN) algorithm became extremely good at identifying things. A 665 imaged dataset is created with labels. With the use of data augmentation techniques such as blurring, rotating, and flipping, the 665 image dataset was transformed into a dataset containing 1995 images. The images are then utilized for training using the Faster-RCNN algorithm with different feature extractors and compared with YOLO v5 of scales small and medium. Faster-RCNN and YOLO v5 architectures are used in object detection algorithms, there will always be a trade-off between latency and accuracy. On the basis of mAP (mean Average Precision), the final results are compared. YOLOv5 had an accuracy of 82% and a Recall of 71.5%, whereas Faster-RCNN had a precision of 63.9% and a Recall of 75%. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

An autoencoder-based deep learning model for solving the sparsity issues of Multi-Criteria Recommender System 2

An autoencoder-based deep learning model for solving the spa...

引用

2nd International Conference on Machine Learning and Data Engineering, ICMLDE 2023

作者： Rajput, Ishwari Singh Tewari, Anand Shanker Tiwari, Arvind Kumar Department of CSE National Institute of Technology Patna Patna800005 India School of Computing Graphic Era Hill University Haldwani263139 India Kamla Nehru Institute of Technology Sultanpur228118 India

In recent times, recommender systems have acquired significant popularity as a solution to the issue of information overload. These systems offer personalised recommendations to users. The superiority of multi-criteria recommender systems over their single-criterion counterparts has been demonstrated, as the former are able to provide more precise recommendations by taking into account multiple dimensions of user preferences when rating items. The prevalent recommendation technique, matrix factorization of collaborative filtering, is hindered by the data sparsity problem of the user-item matrix. On the other hand, it is noteworthy that deep learning techniques have demonstrated significant potential in various research domains, including but not limited to image processing, pattern recognition, computer vision, and natural language processing. In recent times, there has been a surge in the utilisation of deep learning techniques in recommender systems, yielding promising outcomes. This study presents a novel approach to multi-criteria recommender systems through the utilisation of deep learning algorithms to mitigate the data sparsity issue. Specifically, deep autoencoders are utilised to uncover complex, non-linear, and latent relationships between users' multi-criteria preferences followed by matrix factorization technique, ultimately leading to more precise recommendations. The proposed model is evaluated by conducting the experiments on the multi-criteria dataset of Yahoo! Movies. According to the outcomes, the proposed approach outperforms the state of the art recommendation methods by generating more accurate and personalized recommendations. Also, it reduces the data sparsity up to 11% from the multi-criteria dataset. © 2024 Elsevier B.v.. All rights reserved.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

image Classification With The Hybrid PvCN 9

Image Classification With The Hybrid PVCN

引用

9th IEEE International Conference on Smart Structures and systems, ICSSS 2023

作者： Pranaya, B. Kumar, A. vinay Pruthviraj, C. Gangothri, N. Chatterjee, Priyadarshini B v Raju Institute of Technology Computer Science and Engineering Telangana Gajwel India B v Raju Institute of Technology Computer Science and Engineering Telangana Kodad India B v Raju Institute of Technology Computer Science and Engineering Telangana Sangareddy India B v Raju Institute of Technology Computer Science and Engineering Telangana Medak India B v Raju Institute of Technology Computer Science and Engineering Telangana Hyderabad India

ISBN: (纸本)9798350384208

Our research introduces an innovative methodology that seamlessly integrates ResNet-50, an advanced Convolutional Neural Network (CNN), with the HalfUNet architecture, enriched by the attention-gate mechanism, to address the complex task of classifying color and black-and-white images with exceptional precision. Our primary objective is to achieve highly accurate classification while effectively discerning between color and black-and-white images. At the core of our approach lies meticulous data preprocessing, incorporating advanced denoising techniques that substantially enhance the quality of our dataset, yielding an impressive Peak Signal-to-Noise Ratio (PSNR) of 71.35%. This PSNR score substantiates the superior quality of our refined dataset and ensures the consistent performance of our model. Leveraging state-of-the-art deep learning techniques and harnessing the unique capabilities of the Half-UNet architecture, particularly its image classification process, enhanced by the attention-gate mechanism, our approach attains an outstanding image classification accuracy rate of 91.3%. This remarkable achievement underscores the robustness and adaptability of our model, enabling it to perform effectively even when confronted with unfamiliar colorization algorithms during testing and limited prior training data. Our research offers an innovative solution that combines ResNet-50 with Half-UNet, incorporating with 1000 images of rigorous dataset pre-processing, quality assessment, and the attention-gate mechanism to achieve remarkable accuracy in image categorization and colorization tasks. © 2023 IEEE.

关键词： image classification

来源：评论

学校读者我要写书评

暂无评论

Deep Background Matting 7th

Deep Background Matting

引用

7th International Conference on Smart City Applications, SCA 2022

作者： El Abdellaoui, Saîd Kachbal, Ilham LAPSSII Technology High School Cadi Ayyad University B.P. 89 Safi Morocco Faculty of Sciences Mohammed V University - Agdal Rabat Morocco

ISBN: (纸本)9783031268519

Matting methods currently play a vital role in image processing and computer vision. In image matting, the most important step is picking specific elements from an image that will be embedded in a different background. Matting results can vary depending on an image’s complexity, making it an ill-posed problem. In this paper, we present a deep learning method that solves the issue of poor performance when the foreground and background colors are similar, which the current methods suffer from. The encoder-decoder algorithm generates an alpha matte and foreground using the image itself, its soft segmentation and motion cues. These results are fed into a Generative Adversarial Network (GAN) in order to improve the matting process. Our proposed method achieves the performance levels of state-of-the-art algorithms, which makes it better than previous methods. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

A combined deep CNN-lasso regression feature fusion and classification of MLO and CC view mammogram image

引用

INTERNATIONAL JOURNAL OF SYSTEM ASSURANCE ENGINEERING AND MANAGEMENT 2024年第1期15卷 553-563页

作者： Sridevi, v. Samath, J. Abdul PSG Coll Arts & Sci Dept Comp Sci Udumalpet India Govt Arts Coll Udumalpet India Chikkanna Govt Arts Coll Dept Comp Sci Tirupur India

Breast cancer is the most frequent disease among women, and it is a serious threat to their lives and well-being. Due to high population expansion, automatic mammography detection has recently become a critical concern in the medical industry. The emergence of computer-assisted systems has aided radiologists in making accurate breast cancer diagnoses. An automated detection and classification system should be implemented to prevent breast cancer from spreading. Breast densities vary widely among women, which causes missed cancers. In the case of breast density, the deep CNN algorithms can significantly reduce radiologist workload and improve risk assessment. The goal of this paper is to offer a deep learning strategy for identifying MLO and CC views of breast cancer as malignant, benign, or normal using an integration of deep convolutional neural networks (CNN) and feature fusion of LASSO (Least Absolute Shrinkage and Selection Operator) regression. The proposed method comprises pre-processing, data augmentation, feature extraction, feature fusion, and classification. The generated features were fed into LASSO regression for the best prediction in this system, which utilized CNN for feature extraction. The fused features were then transferred to CNN's fully connected layer for mammography classification. In our experiment, the publically available dataset CBIS-DDSM (Curated Breast Imaging Subset of DDSM) was employed. The proposed work gained an accuracy of 99.2%, specificity of 98.7%, AUC of 99.8%, sensitivity of 99.4%, and FI-score of 98.7%, which is higher than multi view CNN without a feature fusion based system.

关键词： CNN LASSO Mammogram cancer classification Regression Feature fusion Breast cancer

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：