Methodology for obtaining textual information from images and its analysis

Abstract

Abstract The paper proposes a solution of the issue of extracting textual information from a document image. There are many interesting solutions of this problem thanks to practical applications, such as text recognition, translation, forensics, etc. The quality of the text information of documents converted into images depends on the type of scene (simple or complex), type of digital camera (low-cost devices, mobile devices), the magnitude and direction of illumination, type of substrate (paper, plastic). These points lead to the fact that the automatic extraction of text from the image is extremely difficult.