Econtechmod

Search Results

Now showing 1 - 4 of 4

Building computer vision systems using machine learning algorithms
(Commission of Motorization and Energetics in Agriculture, 2017) Boyko, N.; Sokil, N.; Lviv Polytechnic National University
In this paper theoretic aspects of machine learning system in the field of computer vision is considered. There are presented methods of behavior analysis. There are offered tasks and problems associated with building systems using machine learning algorithm. The paper provides signs of problems that can be solved by using machine learning algorithms There is demonstrated step by step construction of computer vision system. The paper provides the algorithm of solving the problem of binary (two classes) classification for demonstration the machine learning algorithm possibilities in image recognition field, which can recognize the gender of the person on the photo. Aspects related to the search of data processing are also considered. There is analyzed the search of optimal parameters for algorithms. An interpretation of results in machine learning algorithm is provided. Binarization methods in machine learning algorithm are offered. There is analyzed the technology for improving the accuracy of machine learning algorithm. There are proposed ways to improve computer vision system in neural systems. Also there are analyzed large software modules that work using machine learning systems. The article provides prospects of powerful information technologies, which are necessary for the proper data selection in learning and configuration of feature extraction algorithm to create a computer vision system.
Use of a cloud storage for implementation informational processes
(Commission of Motorization and Energetics in Agriculture, 2017) Boyko, N.; Lviv Polytechnic National University
In this paper, description of a concept of cloud storage is offered. One of the most practical methods of storing required information to cloud storage is also considered. Method of creating screenshots is proposed. Theoretical research is carried out and there are justified advantages and disadvantages of different methods of storing information in social networks. The best technique of creating screenshots has been practically implemented. Problems that might come up while working with approach given in this article and its solutions are stipulated. There is analyzed a necessity of setting a zoom parameter which is to transfer after transmitting size value of the picture in social networks. In the article the parameter that specifies the width of the final image and clearly affects the quality of the image is also considered. There is analyzed the effectiveness of creating an information system that saves time for such information processes as tracking the photo and its comments. In the paper the task of changing bets, which are not immediately fixed in social networks is optimized. Also there is implemented in practice a scalability problem of information processes in social networks. In the article a separation of the script is also put into practice. One part of which directly performs the request of image and downloads it to cloud storage. With help of another part the information process is transmitted on the photo where the user identifies.
Basic concepts of evolution in agents calculating and agents system
(Commission of Motorization and Energetics in Agriculture, 2016) Boyko, N.; Kutyuk, O.
The basics concepts of evolution in agents calculating are discovered in this work and are showed their directions and applications. Before explaining what is agent and its description, there were given a bit of its history and the difference between agents and programs. Were given basic types of agents on examples and figures. The main task of agents is to require a large number of interactions for which most mathematical modeling methods are unsuitable. Were analyzed agent systems architecture and a description of their main parts. Principles of work with mobile and intelligent agents are considered. Furthermore, were exemplify the reasons and situations of use either intelligent agents or mobile agents. Also, their examples were showed on different examples and figures. Technology and application tools which uses in the process are represented. Analysis of JADE-technology are carried out. On the market today there are analogues of JADE, but most of the systems are relatively new and require many improvements, some are under development prototypes. Also, were given description of main tools and features of JADE. It will help a lot in elaboration of agents. Advantages and disadvantages of using agent approach are showed for creating system of data processing and they show their versatility compared with other systems.
Basic concepts of dynamic recurrent neural networks development
(Commission of Motorization and Energetics in Agriculture, 2016) Boyko, N.; Pobereyko, P.
In this work formulated relevance, set out an analytical review of existing approaches to the research recurrent neural networks (RNN) and defined precondition appearance a new direction in the field neuroinformatics – reservoir computing. Shows generalized classification neural network (NN) and briefly described main types dynamics and modes RNN. Described topology, structure and features of the model NN with different nonlinear functions and with possible areas of progress. Characterized and systematized well-known learning methods RNN and conducted their classification by categories. Determined the place RNN with unsteady dynamics of other classes RNN. Deals with the main parameters and terminology, which used to describe models RNN. Briefly described practical implementation recurrent neural networks in different areas natural sciences and humanities, and outlines and systematized main deficiencies and the advantages of using different RNN. The systematization of known recurrent neural networks and methods of their study is performed and on this basis the generalized classification of neural networks was proposed.

Econtechmod

Browse

Filters

Settings

Sort By

Results per page

Search Results