检索结果-内蒙古大学图书馆

Anomaly-Focused Augmentation Method for Industrial Visual Inspection

IEEE Access 2025年 13卷 89118-89139页

作者： Melvinlara, Melvin A. Hameed, Ibrahim A. Koziorek, Jiri Faculty of Electrical Engineering and Computer Science Department of Cybernetics and Biomedical Engineering Ostrava70800 Czech Republic Department of Mechanical Engineering and Technology Management Ås1433 Norway

In industrial inspection, the detection of surface defects - such as scratches, dents, or other defects - is crucial for ensuring product quality. However, the limited availability of annotated images of such defects poses challenges for developing reliable detection models. This paper proposes a novel image augmentation algorithm to address this limitation by generating synthetic imperfections from a small sample set. Unlike traditional approaches that augment entire images, the proposed approach isolates defects and applies spatial transformations (e.g., rotation, shear) before blending them into background images. This allows for the creation of diverse training datasets. The algorithm employs a 2D binary map to encode the spatial relationship between defect and background regions of images and then applies an AI-driven model to generate new realistic blended images. By leveraging a generator-discriminator framework, the augmentation process is iteratively refined during training to maintain the style and texture of the original imperfections. Experimental results show that the proposed method significantly enhances the performance of defect detection models compared to existing augmentation techniques while preserving the realism of the generated images. © 2013 IEEE.

关键词： Photointerpretation

来源：评论

学校读者我要写书评

暂无评论

A Real-Time Automatic Structural-Loss Detection and Stopping Rule of Semiconductor Single-Crystal-Silicon-Growth and

引用

IEEE Transactions on Semiconductor Manufacturing 2024年第3期37卷 304-315页

作者： Song, Wheyming Tina Liao, Yu-Fan Feng Chia University Department of Information Engineering and Computer Science Taichung407 Taiwan National Tsing Hua University Department of Industrial Engineering and Engineering Management Hsinchu300 Taiwan

The occurrence of the 'structural-loss' defect during single-crystal silicon growth (SCSG) is a significant issue in semiconductor manufacturing. When structural-loss occurs, it signifies a deviation from the desired quality of single-crystal formation, leading to the need to halt the growth process. Currently, there is a lack of scholarly literature addressing the determination of an optimal stopping time to promptly halt the process upon the occurrence of the defect on-line. Our research makes a substantial contribution by addressing this gap in the SCSG process, specifically focusing on orientations and . The study utilizes advanced AI with YOLO-v7 and innovative approaches. These include precise annotation of crystal misorientation features through a comprehensive definition of structural-loss and novel labeling techniques, identification of optimal hyper-parameters through a robust design, and the implementation of effective stopping rule mechanisms. Significant progress has been achieved in decision-making through the implementation of the stoping time shift to terminate the SCSG process within an average of less than 3 minutes for orientations (with a standard error of 0.3 minutes) and less than 5 minutes for orientations (with a standard error of 0.5 minutes). The promising results indicate that the proposed approaches have the capability to substitute manual inspections, opening up possibilities for new perspectives in this particular field. © 1988-2012 IEEE.

关键词： Single crystals

来源：评论

学校读者我要写书评

暂无评论

Text recognition using improved dual attention based on textual double embedding network with aquila optimization algorithm

引用

International Journal of Information Technology (Singapore) 2024年 1-15页

作者： Singhdev, Harsiddhi Gupta, Shruti Srivastava, Vivek Saxena, Apoorva Department of Computer Science and Engineering ABES Engineering College Uttar Pradesh Ghaziabad India Department of Computer Science and Engineering Axis Institute of Technology & amp Management Uttar Pradesh Kanpur India

Text recognition is the process that changes an image of text into a system readable text format. Different approaches were suggested related to text detection but the existing methods accuracy is low and error rate is high. Therefore, a text recognition using Improved Dual-attention based on Textual Double Embedding (IDTDE) with aquila optimization algorithm is proposed in this manuscript for effective text recognition. In this method the input image is taken from several datasets. Usually, the images have some kind of noise in it, and to eliminate that, the Structural interval gradient filtering preprocessing technique is used. Then, the ternary pattern and discrete wavelet technique is use to extract best features. Next, residual-based temporal attention convolutional neural network is utilized for the text classification and character identification are effectively attained by and IDTDE network. The Aquila optimization algorithm is used to optimize the network for best extracted text as output. Experimental outcomes demonstrate that the introduced approach accomplishes the high accuracy rates of 98.14%, 98.89%, and 90.47% on the IIIT5K, ICDAR 2013, and ICDAR 2015 datasets, respectively, surpassing the performance of existing frameworks. © Bharati Vidyapeeth's Institute of computer Applications and management 2024.

关键词： Aquila optimizer Character recognition Gradient filtering Improved self-attention Text recognition

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis of Movie Reviews Using Machine Learning

Sentiment Analysis of Movie Reviews Using Machine Learning

引用

2025 International Conference on Pervasive Computational Technologies, ICPCT 2025

作者： Yadav, Avantika Garg, Payal Jain, Tarun Manipal University Department of Computer Science & Engineering Jaipur India Gl Bajaj Institute of Technology and Management Department of Computer Science & Engineering Greater Noida India

ISBN: (纸本)9798331508685

This paper delves into the critical role that machine learning plays in the accurate classification of sentiments expressed in movie reviews. Throughout the study we research and apply NLP using deep learning and machine learning algorithms to train and test our IMDb 50K movie reviews dataset on different models and feature embedding to compare their accuracies and choose the most optimal model for sentimental analysis. Ultimately, using deep neural network model BERT neural gives us the best accuracy of 94.2% compared to lexicon-based approaches due to the limitations in feature extractions like Bag of Words and TF-IDF. © 2025 IEEE.

关键词： Bag of Words BERT Deep Learning IMDb Dataset Machine Learning Natural Language Processin Sentiment Analysis TF-IDF Training and Testing

来源：评论

学校读者我要写书评

暂无评论

Skin Lesion Segmentation Using Deep Learning 2nd

Skin Lesion Segmentation Using Deep Learning

引用

2nd International Conference on Cyber Intelligence and Information Retrieval, CIIR 2023

作者： Sarkar, Kaustav Haralalka, Varun Shaw, Vishal Singh, Rishav Raj Jaiswal, Rayirth Pramanick, Tridip Department of Computer Application and Science Institute of Engineering and Management Kolkata India

ISBN: (纸本)9789819735938

Dermoscopic image analysis has gained significant importance for dermatological applications due to its potential in eliminating observer bias. Segmenting lesion areas automatically in these images is crucial for expedited disease diagnosis. In this study, an innovative deep learning-based approach for precise lesion delineation has been proposed. Leveraging a deep convolutional neural network, where input images are included after appropriate preprocessing. Pedro Hispano Hospital (PH2) 2013 dataset has been employed, comprising dermoscopic images and lesion binary masks as ground truth for evaluation. The approach uses the SegNet architecture, which achieves remarkable accuracy of 99.14%, surpassing benchmark standards (DC 96.94, JI 99.11).This research offers a promising solution for automated lesion segmentation in dermoscopic images, advancing the field of medical image analysis. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Image analysis

来源：评论

学校读者我要写书评

暂无评论

An Analytical Survey on Credit Scoring Mechanism in India 2nd

An Analytical Survey on Credit Scoring Mechanism in India

引用

2nd International Conference on Cyber Intelligence and Information Retrieval, CIIR 2023

作者： Bhowmik, Ankan Basu, Arkajyoti Basu, Arittra Paul, Anish Roy, Atrri Department of Computer Application and Science Institute of Engineering and Management Kolkata India

ISBN: (纸本)9789819735938

Credit score is an important indicator to lenders of your capability to repay loans. A consumer's creditworthiness is indicated by a number between 300 and 850 called a credit score. The advanced the score, the better a borrower looks to implicit lenders. It is predicated on data gathered from lenders and is consolidated in the Credit Information Report or CIR. This paper reviews to check one’s report regularly, also helps you to identify any disagreement that may be lowered your score, and helps to recover the lower score to advanced situations score. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Credit ratings Credit score Loan Net profit

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Machine Learning Models in Network Intrusion Detection Systems 2

Performance Analysis of Machine Learning Models in Network I...

引用

2nd International Conference on Computational Intelligence, Communication Technology and Networking, CICTN 2025

作者： Sharma, Pakhi Dhanda, Namrata Yadav, Vandana Computer Science & Engineering Department Amity University Uttar Pradesh Lucknow India Computer Science & Engineering Department SR Institute of Management and Engineering Uttar Pradesh Lucknow India

ISBN: (纸本)9798331530389

Networks plays a vital role in modern society and cyber security has become a critical area that needs to be acknowledged. A crucial part of cybersecurity frameworks intended to track network traffic and spot suspicious activity, and malicious activity is an intrusion detection system(IDS). Integrating machine learning (ML) algorithms into IDS greatly improves the system's ability to study complex network traffic and accurately identify abnormal patterns. In addition, the incorporation of algorithms reduces the complexity of large data sets and the training time of models. In this paper, we propose the application of three algorithms Decision Tree(J48), SVM, and Naïve Bayes on the NSL-KDD dataset. These algorithms are applied after preprocessing dataset, which includes feature selection and class imbalance resolution. After applying the aforementioned algorithms, we assessed their performance based on metrics such as confusion matrix, accuracy. The results of the environment clearly indicate which algorithm works best with the data set and achieves high accuracy with minimal time required to build the model. © 2025 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

An Extensive Review on Deep Learning Based Approaches for Brain Tumor Classification

An Extensive Review on Deep Learning Based Approaches for Br...

引用

Congress on Smart Computing Technologies, CSCT 2023

作者： Ghosh, Debashis Uma Devi, G. Department of Computer Science and Engineering University of Engineering and Management Rajasthan Jaipur303 807 India

ISBN: (纸本)9789819780952

An accurate and early brain tumor identification and classification system is essential for making a proper and on-time treatment decision. Manually, brain tumors can be diagnosed and classified by analysis of histopathological reports of biopsy samples, but it is time-consuming and has a high chance of wrong identification and classification, which may put the patient in danger. Recent research and developments in Artificial Intelligence, Machine Learning, and Deep Learning have started to play important roles in medical diagnosis and support doctors in making smart treatment decisions. Convolutional Neural Networks (CNNs) are a powerful deep learning-based technique that has recently been applied to the automatic detection and classification of brain tumors. In this literature review, we have reviewed the research papers related to automatic brain tumor identification and classification approaches based on deep learning. A lot of research based on DL has been done in the last few years to implement brain tumor classification and segmentation models with significant results. Some of the drawbacks may be found in various previously proposed methods. So, more improved deep learning models may be implemented in further research by comparing, changing, and combining the various previously proposed models. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhanced Convolutional Neural Network for Sign Character Recognition with Mixed Pooling and Mish Activation Function

Enhanced Convolutional Neural Network for Sign Character Rec...

引用

2024 IEEE International Conference on Intelligent Systems and Advanced Applications, ICISAA 2024

作者： Sur, Hritajit Mandal, Gourab Das, Koushik Nandia, Apurba Das, Avik Kumar Ghosh, Arijeet Institute of Engineering and Management University of Engineering and Management Cybersecurity & Blockchain Technology Department of Computer Science Engineering Kolkata India

ISBN: (纸本)9798331539948

Communication is an essential part of our daily lives, especially for deaf and dumb people who face difficulty in communicating with others. This paper introduces a method of using sign languages, i.e., to convert various hand gestures to appropriate text. The proposed method will help people with disabilities communicate or express their thoughts and opinions to others. By offering cutting-edge methods for sign character recognition, this neural network model aims to close the communication gap and enhance the quality of life for those it serves. In this work, a dataset of hand gestures is divided into training, validation, and testing sets to apply a convolutional neural network (CNN) that has been used to properly recognise the characters on the image. The CNN is further tested on random test data to obtain a reliable interpretation of the sign language we want. The uniqueness of this model lies in the fact that it makes use of mixed pooling operation and mish activation function which train the data more smoothly and in a strong, robust way, which overall improves the accuracy and precision in recognising sign characters. © 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

Exploring Machine Learning Approaches for Analysis and Prediction in the Indian Stock Market 15

Exploring Machine Learning Approaches for Analysis and Predi...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Buche, Arti Chandak, M.B. Department of Computer Science and Engineering Shri Ramdeobaba College of Engineering and Management Nagpur India

ISBN: (纸本)9798331300579

Predicting the Indian stock market with accuracy is a considerable challenge, mainly due to its nonlinear time series nature. Nevertheless, the advent of diverse Machine Learning methodologies has sparked an extensive investigation in this domain. Primarily, scholarly inquiries have concentrated on harnessing Supervised Machine Learning algorithms for predicting the prices of individual stocks and forecasting movements in stock indices. Furthermore, researchers have explored the utilization of Unsupervised Machine Learning algorithms and statistical approaches to anticipate stock prices. Another area of exploration involves constructing portfolios that maximize profits while minimizing risks. Moreover, the influence of social media, which generates an abundance of data, has become a crucial factor impacting experts' decisions to buy or sell stocks. This survey aims to identify the various approaches researchers employ to improve the accuracy of machine learning algorithms in analyzing the Indian Stock Market. It highlights the importance of taking into account all pertinent aspects of the stock market prior to making predictions. © Grenze Scientific Society, 2024.

关键词： Adversarial machine learning

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：