In this article, a system of handwritten or printed text recognition in the image has been developed. Empirical methods of image processing and statistical models of machine learning and simulation are being developed in two directions: the detection of text on the image and the recognition of the text. Thus, in this paper, algorithmic software tools that combine these two areas in the software created for the operating systemiOS 11.0 or later for devices of the company Apple – iPhone, iPad that support this operating system are developed.
- Paramud Y., Yarkun V. Algorithmic and software means of handwritten symbol recognition // Bulletin of the National University “Lviv Polytechnic” “Computer Systems and Networks” No 881, Lviv, Ukraine. 2017. – P. 98–106.
- Yarkun V., Paramud Y. Algorithmic and software synchronization of information exchange // Bulletin of the National University “Lviv Polytechnic” “Computer Systems and Networks” No 857, Lviv, Ukraine. 2016. – P.111–118.
- Datong Chen, Jean-Marc Odobez, Herve Bourlard. Text detection and recognition in images and video frames // Pattern Recognition journal, 2003. – P. 595–608.
- Convolutional Neural Network [Electronic resource]/ Stanford. – Access mode: http://ufldl.stanford.edu/tutorial /supervised/ConvolutionalNeuralNetwork/.
- Convolutional networks [Electronic resource]/ Stanford. – Access mode: http://cs231n.github.io/convolutionalnetworks/#overview.
- Сoreml [Electronic resource]/ Apple. – Access mode: https://developer.apple.com/documentation/coreml.
- Bachelor’s degree dissertation [Electronic resource]/Github. – Access mode: https://github.com/VitaliyYarkun/Bachelor-s-degree-dissertation.
- NIST. American national standard for information systems – data format for the interchange of fingerprint, facial, and, scar mark and tattoo (smt) information, ansi-itl 1-2000 (nist special publication 500–245), September 2000.