检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Kowaleczko, Pawel Tarasiewicz, Tomasz Ziaja, Maciej Kostrzewa, Daniel Nalepa, Jakub Rokita, Przemyslaw Kawulok, Michal KP Labs Gliwice Poland Warsaw University of Technology Warsaw Poland Silesian University of Technology Faculty of Automatic Control Electronics and Computer Science Gliwice Poland

Insufficient image spatial resolution is a serious limitation in many practical scenarios, especially when acquiring images at a finer scale is infeasible or brings higher costs. This is inherent to remote sensing, including Sentinel-2 satellite images that are available free of charge at a high revisit frequency, but whose spatial resolution is limited to 10 m ground sampling distance. The resolution can be increased with super-resolution algorithms, in particular when performed from multiple images captured at subsequent revisits of a satellite, taking advantage of information fusion that leads to enhanced reconstruction accuracy. One of the obstacles in multi-image super-resolution consists in the scarcity of real-world benchmarks—commonly, simulated data are exploited which do not fully reflect the operating conditions. In this paper, we introduce a new MuS2 benchmark for super-resolving multiple Sentinel-2 images, with WorldView-2 imagery used as the high-resolution reference. Within MuS2, we publish the first end-to-end evaluation procedure for this problem which we expect to help the researchers in advancing the state of the art in multi-image super-resolution. Copyright © 2022, The Authors. All rights reserved.

关键词： Remote sensing

来源：评论

学校读者我要写书评

暂无评论

Real-time testing of vision-based systems for AGVs with ArUco markers

Real-time testing of vision-based systems for AGVs with ArUc...

引用

IEEE International Conference on Big Data

作者： Katarzyna Filus Łukasz Sobczak Joanna Domańska Adam Domański Rafał Cupek Polish Academy of Sciences Institute of Theoretical and Applied Informatics Gliwice Poland Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice Poland

ISBN: (纸本)9781665480468

Automated Ground Vehicles (AGVs) use deep-learning-based vision systems to perceive the surrounding environment and extract relevant information about it. Although deep learning models offer high capabilities, they require large amounts of data to be properly trained and tested. Testing is especially important when off-the-shelf models are used by the AGVs - to examine whether they can meet the demands of complex environments such as the production halls of automated factories. One area of such perception algorithms is object recognition. To test such systems, we propose a solution based on ArUco fiducial markers used for automatic labeling of objects. Our solution can be used to test deep learning systems in real time directly on a robot. Our solution requires minimal interference with the environment and additional infrastructure - the desired objects only need to be marked with a marker printed on a home printer. Therefore, the presented testing procedure can be used for testing of AGVs in real-life environments during a real ride from an actual robot perspective. Data gathered during the online testing can be used for the offline comparison of the accuracy of different deep learning models. Although we focus on the online and offline testing in our study, we also incorporated a marker masking procedure. Therefore, the resulting datasets may also be used for training.

关键词： Deep learning Training Computational modeling Big Data Real-time systems Data models Object recognition

来源：评论

学校读者我要写书评

暂无评论

FQSqueezer: k-mer-based compression of sequencing data (vol 10, 578, 2020)

引用

SCIENTIFIC REPORTS 2020年第1期10卷 1-9页

作者： Deorowicz, Sebastian Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 16 44-100 Gliwice Poland. sebastian.deorowicz@polsl.pl.

The amount of data produced by modern sequencing instruments that needs to be stored is huge. Therefore it is not surprising that a lot of work has been done in the field of specialized data compression of FASTQ files. The existing algorithms are, however, still imperfect and the best tools produce quite large archives. We present FQSqueezer, a novel compression algorithm for sequencing data able to process single- and paired-end reads of variable lengths. It is based on the ideas from the famous prediction by partial matching and dynamic Markov coder algorithms known from the general-purpose-compressors world. The compression ratios are often tens of percent better than offered by the state-of-the-art tools. The drawbacks of the proposed method are large memory and time requirements.

关键词：

来源：评论

学校读者我要写书评

暂无评论

FAST FOURIER TRANSFORM ALGORITHM FOR EXTRACTION OF ATTENUATION, PHASE, AND SCATTERING IMAGES IN PHASE-CONTRAST IMAGING

UPB Scientific Bulletin, Series C: Electrical Engineering an...

引用

UPB Scientific Bulletin, Series C: Electrical Engineering and computer science 2023年第4期85卷 137-150页

作者： Ciobanu, Ionuț-Cristian Safca, Nicoleta Anghel, Elena Rășinar, Cezara Popescu, Dan Faculty of Automatic Control and Computer Science University POLITEHNICA of Bucharest Romania Extreme Light Infrastructure – Nuclear Physics Magurele Romania University POLITEHNICA of Bucharest Bucharest Romania Faculty of Chemical Engineering and Biotechnologies University POLITEHNICA of Bucharest Romania

Phase contrast imaging is based on the refraction of the X-ray beam as it passes through the material and shows excellent results for biological tissues. In this paper, we present a new approach for a Python algorithm that can filtrate phase contrast images (attenuation, phase, and scattering) and improve the identification of the interest object. The images used in this work were obtained with a 5.66 m long Talbot-Lau interferometer with a record angular sensitivity of 0.82 µradians, a phase stepping procedure of 16 steps, and a mean energy of 30 keV. The sample used was 0.89 mm diameter fiber from an accredited mammographic phantom. The proposed algorithm extracts the attenuation, phase, and scattering images from the dataset using Fast Fourier Transform. For filtration, we implemented a non-linear filter (median) to eliminate the noise generated by the acquisition system and the projection processing and to improve the contrast. The results were analyzed by calculation of the contrast-to-noise ratio (CNR). © 2023, Politechnica University of Bucharest. All rights reserved.

关键词： Fast Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

Ranking of attributes—comparative study based on data from stylometric domain

引用

Procedia computer science 2022年 207卷 2737-2746页

作者： Beata Zielosko Urszula Stańczyk Krzysztof Żabiński Institute of Computer Science University of Silesia in Katowice Będzińska 39 41-200 Sosnowiec Poland Department of Graphics Computer Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 2A 44-100 Gliwice Poland

The area of feature selection methods constantly expands along with the development of artificial intelligence domain, and has great impact on almost every field, whenever data is processed and explored. The paper presents research where a ranking method was proposed, inspired by an approach which comes from an algorithm for induction of decision rules. The ranking procedure was based on calculation of standard deviation for attributes, taking into account assigned class labels. This method was compared with another ranking mechanism, a modified version of popular Relief algorithm, with incorporating characteristics of variables by supervised discretisation. Comparison of obtained results included the aspect of knowledge representation as well as the perspective of the accuracy for constructed rule-based classifiers. The experiments were performed on datasets from stylometry domain, where authorship attribution was considered as a classification task, and stylometric descriptors as characteristic features defining writing styles of authors.

关键词： Feature selection Ranking of attributes Relief Standard deviation Decision rules Stylometry Supervised discretisation Authorship attribution

来源：评论

学校读者我要写书评

暂无评论

Decentralized Machine Learning for Face Recognition

Decentralized Machine Learning for Face Recognition

引用

International Symposium on Parallel and Distributed Computing

作者： Ioana Brănescu Radu-Ioan Ciobanu Ciprian Dobre Constandinos Mavromoustakis Faculty of Automatic Control and Computers University Politehnica of Bucharest Romania National Institute for Research and Development in Informatics Bucharest Romania Department of Computer Science University of Nicosia Cyprus

As the fields of machine learning and computer vision are developing, facial recognition systems are becoming increasingly popular and are slowly being widely used in various fields like security, surveillance and medicine. This paper presents the design and development of a facial recognition solution that works in a distributed context, given that the devices used for capturing the images do not have the ability to train models capable of achieving good enough accuracy on large amounts of data. Thus, a method is presented in which the detection of human faces and the characteristics extraction are done locally based on a pre-trained FaceNet model. These characteristics are sent to a strong processing unit where a global model is trained and then transferred back to the clients, where it can be used for recognition. Through experimental evaluation, we show that our solution is efficient and exhibits high accuracy values.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Energy, QoS and Security Aware Edge Services 1

引用

2nd International Symposium on Security in computer and Information sciences, EuroCybersec 2021

作者： Gelenbe, Erol Nowak, Mateusz P. Frohlich, Piotr Fiolka, Jerzy Checinski, Jacek Institute of Theoretical and Applied Informatics Polish Academy of Sciences Baltycka 5 Gliwice44-100 Poland LabI3S Universite Cote d’Azur Grand Chateau Nice06103 France Faculty of Automatic Control Electronics and Computer Science The Silesian University of Technology Akademicka 16 Gliwice44-100 Poland

ISBN: (数字)9783031093579

ISBN: (纸本)9783031093562

With the development of communication technologies and the increasing bandwidth of optical fibres and transmission speeds in current 5G and future 6G wireless networks, there is a growing demand for solutions organising traffic in such networks, taking into account both end-to-end transmissions and the possibility of data processing by edge services. The most pressing problems of today’s computer networks are not only bandwidth and transmission delays, but also security and energy consumption, which is becoming increasingly important in today’s climate. This paper presents a solution based on neural networks that organises network traffic taking into account the above criteria - quality of service (QoS), energy consumption and security. © 2022, The Author(s).

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Based Cascade Filtering System for fNIRS Data Analysis

Machine Learning-Based Cascade Filtering System for fNIRS Da...

引用

2023 Progress in Applied Electrical Engineering, PAEE 2023

作者： Pelc, Mariusz Mikolajewski, Dariusz Ruotsalo, Tuukka Leiva, Luis A. Sudol, Adam Gorzelanczyk, Edward Jacek Lysiak, Adam Kawala-Sterniuk, Aleksandra Opole University of Technology Faculty of Electrical Engineering Automatic Control and Informatics Opole Poland Institute of Computer Science Kazimierz Wielki University Bydgoszcz Poland University of Helsinki Finland University of Copenhagen Denmark Faulty of Natural and Technical Sciences University of Opole Opole Poland University of Opole Faculty of Natural and Technical Sciences Opole Poland Institute of Philosophy Kazimierz Wielki University Bydgoszcz Poland

ISBN: (纸本)9798350316254

This paper presents a preliminary study on the use of machine learning-based methods to select the appropriate parameters of cascade filters in the analysis of brain signals recorded using functional infrared spectroscopy (fNIRS), which shows the level of oxygenation in the brain and, unlike EEG signals (showing electrical brain activity), are less prone to potential interference, disturbances or artifacts occurrence. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Methods and tools for automatic or semi-automatic recognition of selected emotions using machine learning algorithms

Methods and tools for automatic or semi-automatic recognitio...

引用

Progress in Applied Electrical Engineering (PAEE)

作者： Piotr Schneider Dariusz Mikołajewski Anna Bryniarska Magdalena Igras-Cybulska Artur Cybulski Walery Marcinowicz Maciej Janiszewski Aleksandra Kawala-Sterniuk Department of Neuroinformatics and Biomedical Engineering Institute of Computer Science Maria Curie-Sklodowska University Lublin Poland Hishoo Sp. z o.o Gdynia Poland Faculty of Computer Science Kazimierz Wielki University Bydgoszcz Poland Faculty of Electrical Engineering Automatic Control and Informatics Opole University of Technology Opole Poland

ISBN: (数字)9798331527563

ISBN: (纸本)9798331527570

automatic or semi-automatic recognition of selected emotions using Machine Learning (ML) algorithms requires practical, fast and accurate implementation of the steps involved in acquiring the actual signals (e.g. speech, facial expression, posture and body movements), recording them in the form of an optimal feature vector/matrix, correctly interpreting the above data and reproducing and generalising this process within the ML system. Depending on the accuracy and speed required and the degree of its ‘invasiveness’ (e.g. by attaching sensors to the user’s body), single or multi-modal interpretations of biometric signals associated with different groups of emotions are possible. This group of studies, despite their relative non-invasiveness, requires the caution prescribed by the Artificial Intelligence (AI) Act and RODO Act and the possibility of being replaced (if necessary) by alternatives. This article aims to analyse the opportunities and risks associated with this group of technologies, with a particular focus on directions for further development.

关键词： Emotion recognition Machine learning algorithms Accuracy Video on demand Speech recognition Machine learning Feature extraction Recording Convolutional neural networks Web sites

来源：评论

学校读者我要写书评

暂无评论

Vehicle type recognition based on audio data 58

Vehicle type recognition based on audio data

引用

58th Hawaii International Conference on System sciences, HICSS 2025

作者： Kobiela, Dariusz Hajdasz, Michal Erezman, Mateusz Nurzyńska, Karolina Zaporowski, Szymon Kurowski, Adam Weichbroth, Pawel Department of Software Engineering Faculty of Electronics Telecommunication and Informatics Gdańsk University of Technology Gabriela Narutowicza 11/12 Gdańsk80-233 Poland Department of Multimedia Systems Faculty of Electronics Telecommunication and Informatics Gdańsk University of Technology Gabriela Narutowicza 11/12 Gdańsk80-233 Poland Department of Algorithmics and Software Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Akademicka 16 Gliwice44-100 Poland

ISBN: (纸本)9780998133188

Identifying different vehicle types can help manage traffic more efficiently, reduce congestion, and improve public safety. This study aims to create a classification model that can recognize vehicle types based on the sound of passing vehicles. To achieve this, a database of raw audio files containing 1763 samples from two sources was assembled. The time-domain signals were converted to a time-frequency representation using the short-time Fourier transform to generate Mel Spectrograms. Mel-frequency Cepstral Coefficients (MFCCs) were also generated using the discrete cosine transform. In our experiments we compared these approaches. Since the data was imbalanced we applied online augmentation. Based on the literature review, we chose a Convolutional Neural Network (CNN) classifier because it is particularly well suited for analyzing large datasets due to its automatic feature extraction, parameter sharing and sparsity. The results showed that Mel Spectrograms were more effective for audio data preprocessing in this particular use case, achieving the highest accuracy of 0.875 and the highest f1-score of 0.877 compared to MFCCs. © 2025 IEEE computer Society. All rights reserved.

关键词： Image coding

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：