The subject of the study is the methods and tools for automation of recognition of road signs at the level of software implementation. Detection of road signs is associated with the processing of a significant amount of video data in real time, which requires significant computing power. Therefore, the purpose of the work is to automate the process of recognition of road signs for filling the databases of navigators, which will allow operatively provide drivers with up-to-date information on established road signs. The following tasks are solved: analysis of methods and software for image recognition; development of the search algorithm for characters in the video frame; implementation of the definition of the contour of the sign; realization of a convolutional neural network for recognition of a sign; testing of applied information technology work. Methods are used: convolutional neural networks; Viola-Jones's method for recognizing objects in an image, the Bousting method as a way to accelerate the recognition process with a large amount of information. Results: Different approaches to the identification of symbols on images, various software tools for object recognition, image transformation for optimal fragment are considered. An algorithm for detecting and recognizing the sign is developed. Using the Viola-Jones method, a fast way to calculate the values of attributes using the integral representation of an image is implemented. The recognition process takes place by constructing a convolutional neural network. Features of the layers of the roller network are considered. Schematically illustrated script recognition. The process of interaction of the system with different data sources is represented by a diagram of precedents. The main result is the creation of information technology for the automated recognition of road signs. The algorithm of its work is presented in the form of a sequence diagram. Conclusions. Using the applied application information technology, recognition of road signs is made with an average probability of 88%, which allows automating the process of filling the database of navigators to a large extent, to increase the reliability and productivity of the given process.
Под корпоративной программой укрепления здоровья работников понимают внедренную работодателем программу, включающую комплекс организационных, профилактических, оздоровительных, социальноэкономических и иных мероприятий, направленных на сохранение и укрепление здоровья работников, и повышение их работоспособности и качества жизни путем формирования у них здорового образа жизни и поведения, уменьшающего риски их профессиональному здоровью. Внедрение корпоративных программ укрепления здоровья работников позволяет работодателю как сохранять здоровье работающих в организации (компании), так и повышать ее имидж. Применительно к работающим выделяют профес
The main medico-social and economic features of the organization of public health systems in leading foreign countries and Russia are considered. The priority directions of further exploratory scientific and practical research, including in urology, have been determined.
An analytical study was carried out aimed at reducing the level of disability of employees of industrial enterprises of the country. It has been revealed that the disability of employees can be a consequence of both various accidents and diseases, including professional ones. An action plan is proposed to reduce the values of disability indicators both in the field of labor protection and medical support of employees. Target corporate programs at industrial enterprises are described. The features of personnel support of work to reduce the level of disability of employees are disclosed. The role of telemedicine technologies in the framework of industrial medicine is reflected. The economic and social effectiveness of the application of the best technologies and practices used in the implementation of the main organizational, medical and socio-economic measures to reduce the disability of employees of an industrial enterprise is shown.
The approach to the formalization of the generation of problem situations applicable to the development of tutoring programs consisting of many tasks is considered. The main errors arising during the software-based generation of parameters are specified. Mathematical modeling of parametrical generation algorithms examined by examples of tasks that make up complex tests on mathematics for secondary schools. The parametric generation method proposed in the article allows getting the large quantitative variations in task problem situations. Thereby, every learner will get a personal unique set of tasks. The structure and functionality of web-tests complex consisting of tasks generated via the proposed method are described. The subject of research in the article is the process of computer training in mathematics. The goal is to develop a method for task generation for mathematical disciplines. Tasks. Research and analysis of the set of mathematical problems. Parameterization of each task and development method and algorithms for automated generation parameters with the determination of incorrect combinations of parameters or problem situations that have no solution. Estimation of borders of admissible for approximate answers. Evaluation of the user solution of a single task and a sequence of tasks. The general objective of the work is to make the software product consisting of a sequence of mathematics tasks. The software should have an extended user interface for the graphical presentation of various problem situations in various mathematical topics. The program must be accessible via the Internet. The following results were obtained: developed methods and algorithms of task generation, which provide correct problem situations and unique parameter sets for each user; described the program complex structure and developed the software system of mathematical web-tests provides two levels of difficulty. Conclusion. The scientific novelty lies in the development of the method of task generation for interactive web tests on the mathematics and its computer implementation with the possibility of graphical representation of tasks and checking of tasks correctness.
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.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.