检索结果-内蒙古大学图书馆

23rd International Conference on control Systems and computer science Technologies, CSCS 2021

作者： Craciunescu, Mihai Baicu, Diana Mocanu, Stefan Dobre, Cristian Faculty of Automatic Control and Computer Science Department of Automatic Control and Applied Informatics Bucharest Romania Rinftech Bucharest Romania

ISBN: (纸本)9781665439398

In this paper, a method of calculating the occupancy of a shelf will be presented. A vision pillar composed of two RGB cameras and two ToF depth cameras will be used to scan a shelf and determine the percentage of emptiness for the products. Since the cameras will scan a small section of the shelf at the time, a stitching method will be applied on both depth and RGB stream. In order to identify categories of products, along with their spatial delimitation within a shelf, a segmentation neural network is used. By combining the segmentation output with the depth information, the distance from the pillar to the product can be determined. Since this type of information does not help a human operator or a supervisor in understanding the state of the shelf or in making an appropriate decision, such as restocking, a method for computing the shelf occupancy is proposed. Optimization must be done to the processing algorithms to enable them to run on an embedded platform, which will allow for the implementation of the proposed architecture on a platform deployed in a production environment. All the computation is done on board of the mobile robot without transmitting the data to an external server. © 2021 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Classification of Skin Lesions from Dermatoscopic Images Using Convolutional Neural Networks 24

Classification of Skin Lesions from Dermatoscopic Images Usi...

引用

24th International Conference on control Systems and computer science, CSCS 2023

作者： Oniga, Maria Sultana, Alina-Elena Popescu, Dan Merezeanu, Daniel-Marian Ichim, Loretta University Politehnica of Bucharest Faculty of Automatic Control and Computer Science Bucharest Romania University Politehnica of Bucharest Faculty of Electronics Telecommunications and Information Technology Bucharest Romania 'St. S. Nicolau' Institute of Virology Bucharest Romania

ISBN: (纸本)9798350313390

Skin cancer is one of the most common types of cancer, and it is caused by a variety of dermatological conditions. Identifying abnormalities from skin images is an important pre-diagnostic step to assist physicians in determining the patient's condition. Thus, to aid dermatologists in the diagnosis process, we proposed five CNN-based classification approaches namely ResNet-101, DenseNet-121, GoogLeNet, VGG16, and MobileNetV2 architectures on which the transfer learning process was applied. The HAM10000-N database consisting of 7,120 images, which was obtained from the original HAM10000 dataset through an augmentation process, was used to train the proposed methods. Moreover, the images from the HAM10000-N were pre-processed by removing hair with the DullRazor algorithm. To evaluate and compare the performance of all networks five metrics were calculated: accuracy, precision, recall, and Fl-score. The best results for the seven-class classification of the HAM10000-N dataset were obtained for DenseNet-121 architecture with 87% accuracy, 0.871 precision, 0.87 recall and 0.872 F1-score. © 2023 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Output Tracking control of a Nonlinear System Based on Takagi-Sugeno Fuzzy Model: Generalized Partial Eigenstructure Assignment Approach

引用

IEEE Access 2024年 12卷 18520-18535页

作者： Soltanian, Farzad Shasadeghi, Mokhtar Mobayen, Saleh Skruch, Pawel University of Alberta Department of Electrical and Computer Engineering EdmontonABT6G 2M9 Canada Shiraz University of Technology Electrical Engineering Faculty Shiraz71557-13876 Iran National Yunlin University of Science and Technology Graduate School of Intelligent Data Science Yunlin Douliu64002 Taiwan AGH University of Science and Technology Department of Automatic Control and Robotics Kraków30-059 Poland

This paper introduces an innovative optimal control approach to achieve output tracking while incorporating H2-performance specifications in a specific class of nonlinear dynamics modeled by the Takagi-Sugeno fuzzy model (TSFM). The primary innovation lies in extending partial eigenstructure assignment to TSFM-based nonlinear systems within the framework of sliding mode control (SMC). We propose a two-step methodology for designing optimal sliding surface gains. Initially, optimal state feedback gains are computed for each rule consequence containing a linear subsystem, adhering to predetermined eigenvalues and satisfying H2-performance criteria. Subsequently, using a convex combination, the overall state feedback gain is calculated and utilized to design sliding matrix gains. The sliding matrix gains are then determined by strategically combining previously calculated state-feedback gains in a convex optimization problem. We reframe the output tracking strategy as a stabilization problem using a virtual control input and reformulate the optimization task concerning tracking state errors. This process yields state feedback gains, sliding gains, and the formulation of the virtual control input. The effectiveness of our approach is verified through comprehensive simulations, emphasizing its capability in addressing output tracking challenges within nonlinear systems. © 2013 IEEE.

关键词： Transient analysis

来源：评论

学校读者我要写书评

暂无评论

Applying Touchscreen as Flight control Inceptor: Investigating the Perceived Workload of Interacting with Sidestick and Touchscreen Inceptors 25th

Applying Touchscreen as Flight Control Inceptor: Investigati...

引用

25th International Conference on Human-computer Interaction, HCII 2023

作者： Zhang, Jingyi Li, Wen-Chin Korek, Wojciech Tomasz Safety and Accident Investigation Centre SATM Cranfield University Cranfield United Kingdom Dynamics Simulation and Control Group SATM Cranfield University Cranfield United Kingdom Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice Poland

ISBN: (纸本)9783031480379

The widespread application of touchscreen technology brought a huge potential for future flight deck design with advanced human-computer interactive mode. The integrated flight displays that enable touchscreen input have been widely examined and validated in the aviation industry. The current research proposed an innovative surface control inceptor based on the touchscreen to control aircraft attitude in flight operations. Fifty-six participants were invited to a flight simulator experiment using the touchscreen and sidestick to perform landing tasks in the scenarios without turbulence and with turbulence. Participants’ heart rate variability (HRV) parameters during flight tasks were collected as biomarkers of mental workload. The two-way repeated measures ANOVA results demonstrated significant interactions of participants’ mean R-R interval (RR) and mean heart rate (HR) between inceptor and turbulence. The simple main effect analysis indicated that in the turbulence scenario, participants’ mean HR significantly decreased and mean RR increased using the touchscreen controller compared to the sidestick inceptor. Furthermore, the standard deviation of normal to normal R-R intervals (SDNN) was higher while performing landing tasks based on the touchscreen in turbulence and non-turbulence scenarios. These findings revealed significant decreases in workload during landing operations from both mental and physical aspects. The applicability of the touchscreen-based control surface inceptor in flight operations was confirmed. This provided a new vision of the advanced aircraft flight control system and future flight deck design for next-generation aviation operations. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Aircraft control

来源：评论

学校读者我要写书评

暂无评论

Transformer-based spectro-temporal fusion for Sentinel-2 super-resolution

Transformer-based spectro-temporal fusion for Sentinel-2 sup...

引用

International Conference on Systems, Signals and Image Processing, IWSSIP

作者： Bartlomiej Pogodzinski Tomasz Tarasiewicz Michal Kawulok Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice Poland

Multi-image super-resolution consists in fusing multiple images of the same scene to generate an image of higher spatial resolution. Recently, it has been demonstrated that for remotely sensed multispectral Sentine1-2 images, fusion performed in both spectral and temporal dimensions improves the reconstruction quality. However, the deep networks elaborated for this purpose are quite large and influence of the input data is difficult to interpret. In this paper, we show how to exploit transformer-based image fusion to reduce the number of trainable parameters by 30% and allow for increased interpretability with means of attention rollout. The reported experimental results performed over simulated and real-world data indicate that this does not affect the reconstruction quality, while at the same time the visualization tools may help in developing techniques for input data selection and preprocessing.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient and Parallel Separable Dictionary Learning 27

Efficient and Parallel Separable Dictionary Learning

引用

27th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2021

作者： Rusu, Cristian Irofti, Paul University Politehnica Bucharest Faculty of Automatic Control and Computers Department of Automatic Control and Computers Bucharest Romania University of Bucharest Faculty of Mathematics and Computer Science Department of Computer Science Bucharest Romania

ISBN: (纸本)9781665408783

Separable, or Kronecker product, dictionaries provide natural decompositions for 2D signals, such as images. In this paper, we describe a highly parallelizable algorithm that learns such dictionaries which reaches sparse representations competitive with the previous state of the art dictionary learning algorithms from the literature but at a lower computational cost. We highlight the performance of the proposed method to sparsely represent image and hyperspectral data, and for image denoising. © 2021 IEEE.

关键词： Image denoising

来源：评论

学校读者我要写书评

暂无评论

Building an Interface for the D Compiler Library 20

Building an Interface for the D Compiler Library

引用

20th RoEduNet Conference: Networking in Education and Research, RoEduNet 2021

作者： Nitu, Razvan Staniloiu, Eduard Creteanu, Cristian Rughinis, Razvan Politehnica University of Bucharest Faculty of Automatic Control and Computer Science Bucharest Romania

ISBN: (纸本)9781665413510

The D programming language has been at the forefront of the memory safe programming languages scene. However, its adoption has been hindered by the scarce availability of 3rd party tools that aid software development such as: linters, code analyzers, integrated development envirnments (IDE) etc. The fundamental reason for this absence is the underdeveloped compiler library, that does not offer a flexible, easy to use interface and misses some of the important features such as symbol resolution and scope retrieval. In this work, we aim at improving the D compiler library by defining a proper interface and adding the currently missing features. To understand what are the needs of a well rounded compiler interface, we analyze existing 3rd party tools and extract a series of common needs. Further, we implement these interfaces directly into the source code of the reference D compiler, with no performance loss. Finally, we upgrade existing tools to use the newly defined compiler interface. By doing so, we unify the different implementations of the same logic. This has the advantage that once the compiler is upgraded to a newer version, all the tools that leverage the compiler library will no longer need to be updated. © 2021 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Identifying Persons from Iris Image 13

Identifying Persons from Iris Image

引用

13th International Conference on Electronics, computers and Artificial Intelligence, ECAI 2021

作者： Costache, Alexandru Badescu, Emilia Popescu, Dan Ichim, Loretta University POLITEHNICA of Bucharest Faculty of Automatic Control and Computer Science Bucharest Romania

ISBN: (纸本)9781665425346

Iris recognition is the biometric application of identifying individuals based on the appearance of the patterns contained by their irises. In this paper, we describe an efficient method for iris recognition using Support Vector Machines, comparing images of the iris of a person to images of irises of other persons. The goal is to obtain a fast and reliable method that can be used in access control applications. Experimental results are presented, discussed and compared to those of similar algorithms, pointing out advantages and disadvantages. © 2021 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Security Audit for the D Programming Language 20

Security Audit for the D Programming Language

引用

20th RoEduNet Conference: Networking in Education and Research, RoEduNet 2021

作者： Nitu, Razvan Staniloiu, Eduard Done, Cristian Rughinis, Razvan Politehnica University of Bucharest Faculty of Automatic Control and Computer Science Bucharest Romania

ISBN: (纸本)9781665413510

Memory corruption has been, traditionally, the number one cause for software vulnerabilities. As a consequence, programming languages that offer automated, compile time memory safety checks have been developed, such as D and Rust. However, since programming languages are pieces of software, they also may suffer from vulnerabilities that may be exploited to bypass the memory safety checking algorithm. In this paper, we perform a security audit of the D programming language. Our findings uncover security holes in the D safety checking system. We show that it is possible to escape expired stack pointers which can be used to ultimately execute arbitrary code. In addition, we discuss and implement potential fixes to the discovered issues. © 2021 IEEE.

关键词： computer programming languages

来源：评论

学校读者我要写书评

暂无评论

Improved conditional GAN for aerial image segmentation

Improved conditional GAN for aerial image segmentation

引用

2021 IEEE AFRICON, AFRICON 2021

作者： Dimoiu, Mihai Lorin Popescu, Dan Ichim, Loretta University Politehnica of Bucharest Faculty of Automatic Control and Computer Science Bucharest Romania

ISBN: (纸本)9781665419840

Generative Adversarial Network (GAN) is an algorithmic architecture containing two neural networks, placed against each other to generate new synthetic images and it has been used successfully in image segmentation. The paper analyzes different GAN implementations for segmentation of images acquired by aerial robots in a real context of a rural zone in Romania. To improve the segmentation performance, a new GAN network is proposed by adding a new layer. Data augmentation was done by the following techniques: mirroring, rotation, scaling, gray scaling, blurring, sharpening, etc. Five classes of region of interest are considered: floods, vegetations, buildings, roads, and dry land. GAN implementations were tested on CPU, GPU, and TPU, on individual computing devices and in the cloud. A new layer was added. The performances were analyzed in terms of learning time, operating time, and statistical indicators. The batch size was generally low: batches of 1, 4 or 16 images were used in this paper. The results confirm that the use of batch achieves the best training and generalization performance in terms of computational cost, for a wide range of experiments. © 2021 IEEE.

关键词： Rural areas

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：