Перегляд за автором "Tvoroshenko, I."
Зараз показано 1 - 20 з 34
Результатів на сторінку
Варіанти сортування
Публікація About classification of the methods in design of medical information systems(Vancouver, Canada, 2021) Tvoroshenko, I.; Mahomet, A.Публікація Analysis of existing methods for searching object in the video stream(2020) Tvoroshenko, I.; Zarivchatskyi, R.Публікація Analysis of methods for detecting and classifying the likeness of human features(2021) Tvoroshenko, I.; Koriakin, I.Публікація Application of deep learning methods for recognizing and classifying culinary dishes in images(International Journal of Academic and Applied Research, 2023) Tvoroshenko, I.; Gorokhovatskyi, V.; Kobylin, O.; Tvoroshenko, A.The paper deals with the actual tasks of computer vision – recognition and classification of objects in images. Deep learning methods based on neural networks are proposed as an alternative to existing methods. A new approach for effective recognition and classification of culinary dishes in images has been developed, which involves the use of the capabilities of the TensorFlow deep learning library and the features of the Convolutional Neural Network. A software application for recognizing and classifying culinary dishes has been developed. Testing of the proposed tool showed that 7 complete epochs of model training provided 77% accuracy, which is a fairly good result, given that one of the main problems in recognizing culinary dishes is interclass similarity. The prospect of further research is to create subclasses for existing general classes of dishes.Публікація Application of procedural generation of game content using software algorithms(2020) Tvoroshenko, I.; Almakaieva, A.Публікація Application of video data classification models using convolutional neural networks(International Journal of Academic and Applied Research, 2023) Tvoroshenko, I.; Pomazan, V.; Gorokhovatskyi, V.; Kobylin, O.This paper explores the classification of video data with convolutional neural networks. It discusses how convolutional neural network architecture can do this task with fastness and maximum efficiency. By analyzing the different convolutional neural network models model was proposed that showed great results in this video classification task. The strengths and weaknesses of the neural network model for video classification have been identified, and the prospects for further work have also been outlined.Публікація Classification of Images Based on a System of Hierarchical Features(2022) Daradkeh Yousef Ibrahim; Gorokhovatskyi, V.; Tvoroshenko, I.; Al-Dhaifallah MujahedThe results of the development of the new fast-speed method of classification images using a structural approach are presented. The method is based on the system of hierarchical features, based on the bitwise data distribution for the set of descriptors of image description. The article also proposes the use of the spatial data processing apparatus, which simplifies and accelerates the classification process. Experiments have shown that the time of calculation of the relevance for two descriptions according to their distributions is about 1000 times less than for the traditional voting procedure, for which the sets of descriptors are compared. The introduction of the system of hierarchical features allows to further reduce the calculation time by 2–3 times while ensuring high efficiency of classification. The noise immunity of the method to additive noise has been experimentally studied. According to the results of the research, the marginal degree of the hierarchy of features for reliable classification with the standard deviation of noise less than 30 is the 8-bit distribution. Computing costs increase proportionally with decreasing bit distribution. The method can be used for application tasks where object identification time is critical.Публікація Cluster representation of the structural description of images for effective classification(Computers, Materials & Continua, 2022) Daradkeh Yousef Ibrahim; Gorokhovatskyi, V.; Tvoroshenko, I.; Zeghid, M.The problem of image recognition in the computer vision systems is being studied. The results of the development of efficient classification methods, given the figure of processing speed, based on the analysis of the segment representation of the structural description in the form of a set of descriptors are provided. We propose three versions of the classifier according to the following principles: “object–etalon”, “object descriptor–etalon” and “vector description of the object–etalon”, which are not similar in level of integration of researched data analysis. The options for constructing clusters over the whole set of descriptions of the etalon database, separately for each of the etalons, as well as the optimal method to compare sets of segment centers for the etalons and object, are implemented. An experimental rating of the efficiency of the created classifiers in terms of productivity, processing time, and classification quality has been realized of the applied. The proposed methods classify the set of etalons without error. We have formed the inference about the efficiency of classification approaches based on segment centers. The time of image processing according to the developed methods is hundreds of times less than according to the traditional one, without reducing the accuracy.Публікація Computational Complexity of the Accessory Function Setting Mechanism in Fuzzy Intellectual Systems(WARSE, 2019) M. Ayaz Ahmad; Tvoroshenko, I.; Jalal Hasan Baker; Lyashenko, V.This article proposes a mechanism for processing fuzzy symptoms, as well as all related factors presented in natural language, which made it possible to improve the intellectual decision-making system in the field of cardiology by adjusting membership functions. The computational complexity of solving the problem according to the criterion of time costs is determined, it is close to exponential from the selected calculation accuracy, which indicates the effectiveness and the absence of disadvantages of the proposed mechanism for a given criterion. The software implementation of the mechanism for setting membership functions in a fuzzy intellectual system is implemented in the Python 3.6 object-oriented programming environment. The laboratory operation of a fuzzy intellectual system has confirmed the high reliability of making objective decisions in the medical subject area.Публікація Current state of development of applications for recognition of faces in the image and frames of video captures(2021) Tvoroshenko, I.; Kukharchuk, V.Публікація Development of an application for recognizing emotions using convolutional neural networks(International Journal of Academic Information Systems Research, 2023) Pomazan, V.; Tvoroshenko, I.; Gorokhovatskyi, V.This paper explores the potential of using convolutional neural networks (CNN) for emotion recognition in marketing research of advertising. Discusses how CNN can be trained to accurately recognize emotions conveyed in advertisements and how this information can be used to gain insights into consumer behavior. By analyzing consumer emotions in response to different types of advertising, marketers can better tailor their advertising campaigns to elicit positive emotional responses and improve overall effectiveness. Overall, the paper demonstrates the potential of CNN-based emotion recognition as a valuable tool to optimize advertising strategies and improve consumer engagement.Публікація Development of models of spatial analysis of status of interactive processes of complex systems(Publishing House “Baltija Publishing”, 2019) Tvoroshenko, I.Given that objects of special purpose function in a priori uncertainty, characterized by a fuzzy space of states, which requires new intellectual approaches to increase the reliability of the decisions that are adopted, characterized by functional and territorial distribution, a complex hierarchy of interacting processes. It is necessary to predict certain requirements to the mathematical apparatus, methods of objectoriented modeling and analysis of interacting processes of a complex system.Публікація Development of web applications for remote learning of English(2021) Tvoroshenko, I.; Andrieieva, A.Публікація Features of methods of issuation of key areas and vector recognition of manuscript symbols on the image(2021) Tvoroshenko, I.; Bielinskyi, Y.Публікація Features of software application development for food recognition using deep machine learning methods(2021) Tvoroshenko, I.; Temchur, K.Публікація Handwritten character recognition models based on convolutional neural networks(International Journal of Academic Engineering Research, 2023) Pomazan, V.; Tvoroshenko, I.; Gorokhovatskyi, V.This paper explores the possibility of using various convolutional neural network models to recognize handwritten Arabic characters. It discusses which of the existing convolutional neural network models best suits this task. The results are analyzed based on which conclusions are drawn regarding the effectiveness or ineffectiveness of specific convolutional neural network models. The paper provides an analysis of the efficiency of convolutional neural network models for the task of recognizing handwritten Arabic characters.Публікація Identification of visual objects by the search request(Kyiv-Uzhorod, 2023) Gorokhovatskyi, V.; Tvoroshenko, I.Публікація Image Classification Based on the Kohonen Network and the Data Space Modification(2020) Gorokhovatskyi, V.; Tvoroshenko, I.In this paper, we propose the solution of visual object recognition in computer vision problems using the classification of descriptors of image key points based on the training of Kohonen neural network on the description data of etalons images. According to the results of training within the set of etalons, the image classification method has been improved by defining a specific data space in the form of a statistical center for each etalon. We propose mathematical models for the bitwise analysis of multiple descriptors searching for the centers and the method for convolution of descriptions from multiple descriptors with the determining a posteriori probabilities for the system of bit centers. Methods of data space transformation of description bits are proposed for various options for Kohonen network training, processing and estimation of class centers. The software implementation of the changed classifier was performed as well as the processing time with different options for determining the space of training data was estimated. Experimental researches confirmed the high efficiency of classification preserving sufficient performance and the ability to use proposed methods in real-time applications.Публікація Information technologies of the processing of the spaces of the states of a complex biophysical object in the intellectual medical system health(WARSE, 2019) Asaad Ma. Babker; Abd Elgadir A. Altoum; Tvoroshenko, I.; Lyashenko, V.This article proposes information technologies for processing the state spaces of a complex biophysical object in an intelligent medical system HEALTH, which made it possible to improve the intelligent decision-making system in the field of cardiology by selecting a probable set of diagnoses with the appropriate coefficients of confidence and gradual refinement of it. Software implementation of the state space processing of a complex biophysical object in an intelligent medical system HEALTH implemented in an object-oriented programming environment. Laboratory operation of the hardware and software of the intelligent medical system HEALTH confirmed the high probability of making objective decisions in the medical subject area. It has been determined that an intelligent medical system has been developed HEALTH can be used both in clinical practice by novice specialists and during training of students and doctors in advanced training institutes.Публікація Interactive Geoinformation Three-Dimensional Model of a Landscape Park Using Geoinformatics Tools(IJASEIT, 2020) Ahmad, M. Ayaz; Tvoroshenko, I.; Baker, Jalal Hasan; Kochura, L.; Lyashenko, V.The relevance of work is due to the lack of free three-dimensional models of landscape parks and the urgent need to create such models for informing and barrier-free adaptation of users. The current state of cartographic support of landscape parks is low and does not meet users, management, and employees of parks. The object of research is the territory of the multicomplex. The purpose of the research is to develop a three-dimensional model of the regional landscape park for informing and barrier-free adaptation of residents and guests of the city using geoinformatics tools. The methods used in this research are general scientific (system analysis, modeling, higher mathematics) and special methods of mathematical and geoinformation analysis. The work uses the ArcGIS and SketchUp software platforms to create, manage, integrate, analyze, and distribute spatial data. Methods used: geodesy, topography, geoinformatics. The results obtained in the course of the work are of practical importance, are real and tested by the management of the regional landscape park. An interactive three-dimensional geoinformation model implemented in the ArcScene software environment will provide an opportunity further to manage the territory of the regional landscape park. An interactive geoinformation three-dimensional model ensured a 12% increase in the level of informational and barrier-free adaptation of residents and guests on the territory of the multicomplex. The prospects for further research on the creation of a mobile application for park visitors with the ability to search for the nearest bus departure point at their location are determined.