The methods of processing biomedical images, namely thermal images, are investigated. Algorithms for calculating the temperature and area of the zone of interest in the manual mode operator-computer, as well as in the automatic mode, are specified. Methods of thermal image processing are presented, namely recursive generalized contour preparation and preparation based on histograms of connections. An experimental study of these methods was performed, as well as a comparison of thermal image segmentation methods in manual segmentation modes, using contour preparation-based segmentation, multilevel segmentation based on recursive generalized contour preparation, and automatic segmentation based on connectivity histograms.
Distributed information systems that support scientific and educational activities can work with various information systems. The main goal of creating a distributed information system supporting scientific and educational activities is to accelerate the pace and improve the quality of information exchange in the scientific environment. The paper considers technological methods for constructing models of information systems designed to support scientific and educational activities. The model under consideration is that the developed model of an information system for working with scientific materials should solve the problems of long-term storage of information, organizing data search by attributes, accumulating and replacing metadata. Based on the analysis of typical scenarios of information servers, the tasks that should be solved when organizing an access control system for distributed information resources are formulated. Within the framework of this technology, three access control models are discussed, which differ in the degree of integration of information server functions with the Z39.50 technologies. The creation and support of distributed information systems and electronic libraries that integrate heterogeneous information resources and operate in various software and hardware environments require special approaches to managing these systems. If the resources or data themselves can be managed locally, even for distributed information systems, then the task of managing access to distributed resources cannot be solved within the framework of local administration. The justification of the last thesis can be seen when considering typical scenarios of the information server, which we will describe below
The article presents an approach to organizing scientific portals based on ontologies. Ontology is the information basis of the Internet portal of knowledge, which should provide integration and systematization of scientific knowledge and information resources of a certain subject, as well as meaningful access to them from any "point" of Internet space. The ontology automatically builds a diagram of the portal's internal database and forms for filling it out, organizes navigation through the portal's information space, and ensures that search queries are formulated in terms of the knowledge portal's subject area. The division of the portal's ontology into subject-independent and subject-specific ontologies makes the portal customizable for almost any field of scientific knowledge. This technology allows declarative adjustment of the ontology during the operation of the knowledge portal, which allows you to track the dynamics of the emergence of new knowledge and information resources on the subject of the portal and thus provides support for its relevance and usefulness.
The paper deals with general issues of organizing access to electronic documents in the framework of scientific and educational activities. Large volumes of already existing information, its continuous growth, the heterogeneous nature of storage and distribution, the lack of a unified way of working with it create many difficulties when using it. Awareness of these difficulties, qualitative changes in the field of information technology and telecommunications have led to the need to solve the problem of finding new approaches to the creation of repositories of information resources, their structure, and the development of tools necessary for users. Currently, such approaches are called “digital” or “electronic” libraries. According to the preliminary concept, an intelligent scientific and educational Internet resource will be an information system accessible via the Internet, providing systematization and integration of scientific knowledge, data, and information resources into a single information space, meaningful and effective access to them, as well as support for their use in solving various scientific and educational tasks. Another problem of the organization of effective information support for scientific and educational activities is that, due to its diversity and multidimensional nature, scientific and educational information resources are dispersed on remote pages of many sites and in distributed electronic libraries and archives. To solve this problem, it is necessary to solve the problem of bringing such resources related to one area of knowledge into a single information space, and also, no less important, to support their logical integrity. Without solving these two related tasks, it is impossible to solve the main task – to provide all participants of scientific and educational activities with meaningful access to integrated information resources and means of their analysis. The support of information systems in the field of scientific and educational activities is relevant, since the need for information always exists. In order to satisfy this need, it is necessary to organize access to various resources.
The paper is devoted to machine learning methods that focus on texture-type image enhancements, namely the improvement of objects in images. The aim of the study is to develop algorithms for improving images and to determine the accuracy of the considered models for improving a given type of images. Although currently used digital imaging systems usually provide high-quality images, external factors or even system limitations can cause images in many areas of science to be of low quality and resolution. Therefore, threshold values for image processing in a certain field of science are considered. The first step in image processing is image enhancement. The issues of signal image processing remain in the focus of attention of various specialists. Currently, along with the development of information technology, the automatic improvement of images used in any field of science is one of the urgent problems. Images were analyzed as objects: state license plates of cars, faces, sections of the field on satellite images. In this work, we propose to use the models of Super-Resolution Generative Adversarial Network (SRGAN), Extended Super-Resolution Generative Adversarial Networks (ERSGAN). For this, an experiment was conducted, the purpose of which was to retrain the trained ESRGAN model with three different architectures of the convolutional neural network, i.e. VGG19, MobileNet2V, ResNet152V2 to add perceptual loss (by pixels), also add more sharpness to the prediction of the test image, and compare the performance of each retrained model. As a result of the study, the use of convolutional neural networks to improve the image showed high accuracy, that is, on average ESRGAN+MobileNETV2 – 91 %, ESRGAN+VGG19 – 86 %, ESRGAN+ResNet152V2 – 96 %.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.