AUTOMATED METHODS AND SYSTEMS OF IDENTIFYING IMAGE FRAGMENTS IN DOCUMENT-CONTAINING IMAGES TO FACILITATE EXTRACTION OF INFORMATION FROM IDENTIFICATED DOCUMENT-CONTAINING IMAGE FRAGMENTS

摘要

FIELD: physics.;SUBSTANCE: processing system of document-containing images is intended for: obtaining an image, applying the feature detectors to the obtained image. Each feature detector creates a set of features associated with the detector from the image, for each of one or more document type models; applying the document type model to the resulting image. The document model contains a set of model features, to create a set of summary feature pairs; applying one or more checks to the set of final feature pairs and, if an image fragment containing a corresponding document type model is present in the image, applying a transformation that converts the positions of the model features to the positions on the image to create and store the positions of one or more points of the document-containing image fragment; and to create and save the orientation of the document-containing image fragment.;EFFECT: ensuring the identification of document-containing fragments in the image.;20 cl, 78 dwg