Significant yield challenges are posed by biotic stress on coffee leaves, which has a negative effect the revenue generation of this highly utilized commodity. Numerous studies have proposed techniques for early detection and classification in leaves. In study, we propose technique called extracted feature ensemble (EFE) classifying healthy infected classes. Transfer learning-based convolutional neural networks (CNNs) custom-designed features used to improve performance. Under concept EFE,...

Abstract This article introduces a method for realizing the Gaussian activation function of radial-basis (RBF) neural networks with their hardware implementation on field-programmable gaits area (FPGAs). The results modeling FPGA chips different families have been presented. RBF various topologies synthesized and investigated. component implemented by this algorithm is an network four neurons latent layer one neuron sigmoid using 16-bit numbers fixed point, which took 1193 logic matrix gate...

The object of this study is information systems within the Internet Things. task addressed involves devising an innovative method for detecting and correcting multi-bit data transmission errors in Things networks based on Golay codes. As main result research, a has been devised codes during byte-by-byte block. distinguished by its coding scheme, which calculating 11 control bits one parity bit twelve bytes original message with subsequent mixing using shift operations before to communication...

Analysis of electroencephalography (EEG) signals gathered by brain-computer interface (BCI) recently demonstrated that deep neural networks (DNNs) can be effectively used for investigation time sequences physical actions (PA) classification. In this study, the relatively simple DNN with fully connected network (FCN) components and convolutional (CNN) was considered to classify finger-palm-hand manipulations each from grasp-and-lift (GAL) dataset. The main aim study imitate investigate...

Deep learning has found widespread application in diabetic retinopathy (DR) screening, primarily for lesion detection. However, this approach encounters challenges such as information loss due to convolutional operations, shape uncertainty, and the high similarity between different types. These factors collectively hinder accurate segmentation of lesions. In research paper, we introduce a novel dual-branch U-Net architecture, referred Dual-Branch (DB)-U-Net, tailored address intricacies...

Nowadays, Kubernetes is an advanced container orchestration tool due to its high reliability, scalability and fault tolerance. However, requires a significant number of resources for work. Therefore, ensure the operation in conditions limited resources, lightweight analogues such as MicroKubernetes K3S were created. These platforms provide easier deployment support. In this paper, authors analyze performance metrics actions adding/removing nodes starting/stopping deployments terms resource...

For effective management of cloud data center resources there is a need to develop and implement simple robust methods that take into account the power consumption, user Service Level Agreement compliance, different conditions resource utilization. In this paper, authors analyze possibility application reinforcement learning method management. Due intensive changes workloads utilization problem should be solved dynamically in an online manner. To address such problem, propose dynamic virtual...

Czasopismo Techniczne » 2018 Volume 5 A real time control system for balancing a ball on platform with FPGA parallel implementation

The object of this study is the software and architectural solutions for specialized systems that monitor state potentially hazardous facilities (hereinafter, PHF). problem solved was choice a successful option an solution composition such systems. A change in monitoring PHF necessary because are usually designed on basis principle parametric control over main parameters PHF. Such record achievement pre-critical (or critical) value one several) characterizing object. Therefore, operational...

Cloud data centers are most dynamic systems in a modern digital world. To deliver the high-performance and fault-tolerant IT services to end users effectively it is necessary develop new methods for center resource management while adapting emergence of requirements. In this paper, authors refine evaluate previously proposed method cloud based on reinforcement learning approach. The takes into account power consumption number SLA violations policy. managed by switching physical servers...

Virtual machine consolidation and scheduling influence directly the cloud cost performance. They play an important role in service granting helping to achieve execution efficiency, user Service Level Agreement (SLA) compliance, utilization of resources, energy saving, increasing profit providers. In this paper authors propose Optimization using Simulated Annealing (OSA) algorithm solve dynamic virtual problem. The problem is presented as extension bin-packing configuration system, function...

The object of the study is IOT system for monitoring state objects. problem being solved development an innovative method detecting and correcting data transmission errors in networks Internet Things systems. essence results that a multiple during byte-by-byte block information has been developed. distinguished by original coding scheme, which involves calculation control bits, as well shuffling bits performing bit shift operations. peculiarity any can be distorted when transmitting code...

Modern cloud data centers must effectively deliver a wide range of IT services to clients ensuring preassigned performance and availability. But the rapid growth large-scale results in necessity developing new methods for resource management center. This paper presents hybrid virtual machine consolidation approach that aims at minimization energy consumption SLA violation by applying modified Best Fit Decreasing heuristic initial (VM) placement adopted beam search algorithm manage VM...

A widespread use of the cloud computing paradigm has increased necessity and significance improving management efficiency data centers. Special attention is paid to solving resource problems. Due intensive changes virtual machine (VM) workloads different conditions utilization VM placement migration problems should be solved optimized continuously in an online manner. To address such authors present approach continuous new allocation migration. The also evaluate a particular policy center...

The problem of the rapid creation effective web applications one class using formal means is considered. A conceptual approach to its solution offered based on an analysis peculiarities construction applications. Our defining standard application architecture and selecting components methods in accordance with user requirements. logical system proposed that uses design as a process outputting formula specified according needs user, which defines schemes execution modules system. An important...

The approach to monuments assessment on the basis of fuzzy logic is developed in article. Key elements - creating rules based quasi-statistics and their improvement by means average indicators for sphere application results; constructing membership functions according histograms, taking into account general appearance, ratio mean values peak values; choice most informative indicators; possibility using method various integrated indicators. assessing state can be used spheres human activity.

Today, the IT-infrastructures are widely used in various spheres of life. The advent large Data Center(DC) has caused a problem there resources management. One this aspect is to allocate DCs virtual machines(VMs) and load. Migration opportunity have added complexity problem. In paper, most forecasting models estimated with aim instrument selection for predicting DC VM s migration. methods compared terms they ability predict VMs migration on real data Google cluster environment. Comparison...

The approach to the integration of applications based on mathematical logic and artificial intelligence for World Data Center (WDC) interdisciplinary scientific investigations has been developed in article. Key elements are a multilevel system architecture, formal logical system, implementation interaction intelligent agents. proposed. inference method solution tree recovery mechanism have elaborated. application research stack modern protocols, enabling communication business processes over...

Coherence evaluation of texts falls into a category natural language processing tasks. The texts’ coherence implies the estimation their semantic and logical integrity; such feature text can be utilized during solving multidisciplinary tasks (SEO analysis, medicine area, detection fake texts, etc.). In this paper, different state-of-the-art methods based on machine learning models have been analyzed. investigation effectiveness for Polish has performed. impact text’s features output value...

The problem of quality services providing at predetermined level and maintaining particular parameters IT-infrastructure elements in a given range was considered. Method functioning estimation the information telecommunication systems based on analysis these elements, with use fuzzy neural network ANFIS proposed.