Моделювання системи розпізнавання математичних виразів у графічному форматі представлення даних
Journal Title: Математичне моделювання - Year 2018, Vol 1, Issue 2
Abstract
SIMULATION OF THE RECOGNITION SYSTEM OF MATHEMATICAL EXPRESSIONS IN A GRAPHICAL FORMAT OF DATA PRESENTATION Mikhailutsa O.M., Pozhuyev A.V., Nebesniuk V.O. Abstract With gradual automation of many processes in various areas the priority is training a computer to recognition of a handwritten material, instead of preparation of specially trained people. The authors direct an attention on the problem of recognition of mathematical expressions, presented in graphical format. This is especially useful by search of scientific documents on mathematical formulas, formation of the knowledge base of formulas, carry of formulas to other documents, withdrawal of the semantic information from the formula, integration with computer algebra systems. Mathematical expressions represent bidimentional hierarchical data structure, which precise definition requires the task of the formal list of mathematical symbols in the knowledge base of formulas. The research of text definition systems has shown that traditionally in a basis two approaches to recognition lay: online and offline. The best recognition results in both approaches show such applications that use neural networks. So, the purpose of the research is, firstly, to increase the efficiency of recognition of mathematical expressions due to the classification and clustering of images, and secondly, to improve the existing methods of recognition of mathematical models, in which algorithm possibilities multistage identifications of the image by various methods of an assessment of mathematical expression are added. By analyzing neural networks and conducting theoretical and practical research, a set of research methods was formed, compared and analyzed, and a stack of technologies and development tools was determined, their relative performance, degree of integration, and usability were determined. The algorithm for recognition of the separate handwritten mathematical text and its export to the electronic document of the LaTeX format has been developed. The received results can be used, in particular, to expand the functionality of existing systems for machine translation of the handwritten text into digital format for information sites, as well as to develop more client-oriented systems of assistance to students and schoolchildren. In the further progress it is planned to develop a system of adaptation of the program to the handwriting of the specific person. References [1] Isupov N. S., Kuchuganov A. V. Raspoznavaniye slitnykh rukopisnykh tekstov s ispol'zovaniyem apparata nechetkoy logiki [Recognition of coherent handwritten texts using the apparatus of fuzzy logic]. The Bulletin of Izhevsk State Technical University named after MT Kalashnikov. – 2012. –№ 3. – pp. 125–128. [2] Saparov A. Yu, Beltyukov A. P. Primeneniye regulyarnykh vyrazheniy v raspoznavanii matematicheskikh tekstov [The use of regular expressions in the recognition of mathematical texts]. The Bulletin of Udmurt University. Maths. Mechanics. Computer science. – 2012. – Vol. 2. – pp. 63–73. [3] Saparov A. Yu., Shirobokova I. Yu. Razrabotka pol'zovatel'skogo interfeysa dlya upravleniya protsessom raspoznavaniya rukopisnykh matematicheskikh formul [Development of the user interface for managing the process of recognizing handwritten mathematical formulas]. The Bulletin of Udmurt University. Maths. Mechanics. Computer science. – 2016. – № 1. – pp. 141–152. [4] Lizunov P.P., Biloshchitsky A.A., Chala L.E., Beloshchitskaya S.V., Kuchanskiy O.Yu., Udovenko S.G. Hibrydnyy pidkhid do analizu ta rozpiznavannya matematychnykh formul z metoyu vyyavlennya v nykh podibnostey [A hybrid approach to the analysis and recognition of mathematical formulas in order to identify similarities in them]. Management of complex systems development. – 2016. – No. 27. – pp. 145–155. [5] Lin X., Gao L., Tang Z. Mathematical formula identification in PDF documents in Proc. of ICDAR ’11. Washington, DC, USA: IEEE Computer Society, 2011, pp. 1419–1423. [6] Chandhok C. A novel approach to image segmentation using artificial neural networks and K-means clustering // Intern. J. of Engineering Research and Applications (IJERA). 2012. Vol. 2. Iss. 3. pp. 274–279. [7] Kruglov V.V., Borisov V.V. Iskusstvennyye neyronnyye seti. Teoriya i praktika [Artificial neural networks. Theory and Practice]. – M.: Hotline – Telecom, 2001. – 328 p. [8] Belim S.V., Larionov S.B. Algoritm segmentatsii izobrazheniya s pomoshch'yu iskusstvennoy neyronnoy seti bez ispol'zovaniya drugikh izobrazheniy [Algorithm of image segmentation using an artificial neural network without using other images]. Radio-optics. MGTU them. N.E. Bauman. Electron. journals – 2017. – № 03. – pp. 43–53.
Authors and Affiliations
О. М. Міхайлуца, А. В. Пожуєв, В. О. Небеснюк
ВИЗНАЧЕННЯ СИЛОВОГО ВПЛИВУ ЗМІННОГО СТРУМУ НА МЕТАЛ У ВАННІ УСТАНОВКИ КІВШ-ПІЧ
DETERMINATION OF THE POWER ACTION OF THE ALTERNATING CURRENT ON METAL IN THE POOL LADLE-FURNACE UNIT Krikent I.V. Abstract The work is devoted to theoretical research in the field of secondary steel processing. The expe...
Концептуальна модель системи операційного управління клієнтською базою користувачів послуг
CONCEPTUAL MODEL OF OPERATIVE MANAGEMENT SYSTEMS CLIENT BASIS OF KORISTUVACH SERVICE Grigorenko V.U., Kadilnikova T.M. Abstract At present, the development of operational management of processes and information technolo...
Адаптация формулы Планка для реальных тел на базе экспериментальных исследований
The methodology of experimental data processing, that allows to adapt the Plank radiation formula for creating the equation of heat radiation for the real solids using the correlation function, is described in this artic...
Проектування та розробка клієнт-серверного додатку на основі однорангових мереж
DESIGN AND DEVELOPMENT OF A CLIENT-SERVER APPLICATION BASED ON PEER- TO-PEER NETWORKS Solodka N.O., Liashenko O.A. Abstract A client-server application for data transmission in peer-to-peer networks is designed. It is p...
Моделювання середньозваженої крупності твердого в завантаженні кульового млина рудою і пісками класифікатора
MODELING OF THE MEDIUM-WEIGHTED FINENESS SOLID IN THE LOADING OF THE BALL MILL OF THE ORE AND THE SAND OF THE CLASSIFIER Matsui A.N., Kondratets V.A. Abstract One of the key issues to improve the efficiency of ore prep...