Document Localization and Classification As Stages of a Document Recognition System

N. S. Skoryukinaa,b,**, D. V. Tropina,b,***, J. A. Shemiakinab,****, and V. V. Arlazarova,b,*

a Federal Research Center “Computer Science and Control” of the Russian Academy of Sciences, Moscow, 119333 Russian Federation

b Smart Engines Service LLC, Moscow, 117312 Russian Federation

Correspondence to: * e-mail: vva@smartengines.com
Correspondence to: ** e-mail: skleppy.inc@smartengines.com
Correspondence to: *** e-mail: daniil_tropin@smartengines.com
Correspondence to: **** e-mail: jshemiakina@smartengines.com

Received 20 January, 2023

Abstract—The article is devoted to approaches and methods for analyzing document images, which were developed and used by scientists of the scientific school of V.L. Arlazarov to solve problems of the type definition and localization of documents with a known structure. It describes the principles for building solutions that have emerged as input data have become more complex and performance requirements have stricted. The methods presented in the article demonstrate the scientific path of the school from working with scanned images to photographs and video stream frames, from the most general classes of documents tied to text structure to the strictest ones based on their visual features.

Keywords: scientific school, classification of documents, localization of documents, metric rectification

DOI: 10.1134/S1054661823040430