User interface modeling for convolutional neural network for complex character recognition

In this article, we design a user interface for prototype desktop application using the capabilities of author’s neural network recognizing texts in Japanese written by one two alphabets – katakana or hiragana. During design, UML notation, Use-Case Diagram, was used to build scenarios program, and BPMN notation describe program’s main algorithm. beginning article short versions previous articles were also given basics proposed method preprocessing machine learning data parameters convolutional model including its efficiency against reference EfficientNetB0. work, principles tool base designing software solution defined, algorithms program designed, created.