Computer implemented method for segmenting a binarized document

Computer implemented method for segmenting a binarized document comprising the steps of: extracting a plurality of connected components from the binarized document and discriminating for at least one of the connected components whether said connected component is a text component based on a homogeneity level value, wherein the homogeneity level value is representative of the level of homogeneity within the local region of said connected component, wherein the local region comprises said connected component and at least one connected component adjacent to said connected component, wherein the homogeneity level value is based on at least one value representative of at least one image characteristic parameter determined for said connected component and on at least one value representative of the at least one image characteristic parameter determined for the at least one connected component adjacent to said connected component.

Octrooi-publicatienummer: WO2020225430

Jaar aanvraag: 2020

Jaar toekenning: 2021

Gevalideerd voor IOF-sleutel: Ja

Toegewezen aan: Universitaire Associatie Brussel