Reconnaissance et Extraction de Pièces d'identité

Olivier Augereau (1) , Nicholas Journet (1) , Jean-Philippe Domenger (1)

1 LaBRI - Laboratoire Bordelais de Recherche en Informatique

This article deals with the problem of recognition and location of identity documents within a scanned image. The aim is to detect if an identity document is present in order to locate and extract the region of interest containing it and then to provide the region to an OCR. Initially, a model of an identity document is defined and a set of points of interest are extracted from this model. In a second step, a set of points of interest is extracted from the query image. This set is matched with the model set. This matching is used to calculate the geometric transformation (translation, rotation, zoom) allowing the registration between the model and query image. The position is determined by using the transformation matrix, so the identity can be extracted.

Mots clés: point d'intérêts; FLANN; SURF; RANSAC; recherche de sous-image

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]; Traitement du texte et du document

Olivier Augereau, Nicholas Journet, Jean-Philippe Domenger. Reconnaissance et Extraction de Pièces d'identité. Colloque International Francophone sur l'Écrit et le Document, Mar 2012, Bordeaux, France. pp.179-194. ⟨hal-00708565⟩