Processing

Please wait...

PATENTSCOPE will be unavailable a few hours for maintenance reason on Tuesday 25.01.2022 at 12:00 PM CET
Settings

Settings

Goto Application

1. WO2019002602 - DETECTION OF MANIPULATED IMAGES

Publication Number WO/2019/002602
Publication Date 03.01.2019
International Application No. PCT/EP2018/067692
International Filing Date 29.06.2018
IPC
G06K 9/00 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
G06K 9/46 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
36Image preprocessing, i.e. processing the image information without deciding about the identity of the image
46Extraction of features or characteristics of the image
G06K 9/62 2006.1
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
CPC
G06F 21/32
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
FELECTRIC DIGITAL DATA PROCESSING
21Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
30Authentication, i.e. establishing the identity or authorisation of security principals
31User authentication
32using biometric data, e.g. fingerprints, iris scans or voiceprints
G06K 9/6256
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6217Design or setup of recognition systems and techniques; Extraction of features in feature space; Clustering techniques; Blind source separation
6256Obtaining sets of training patterns; Bootstrap methods, e.g. bagging, boosting
G06K 9/6269
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6267Classification techniques
6268relating to the classification paradigm, e.g. parametric or non-parametric approaches
6269based on the distance between the decision surface and training patterns lying on the boundary of the class cluster, e.g. support vector machines
G06K 9/6277
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6267Classification techniques
6268relating to the classification paradigm, e.g. parametric or non-parametric approaches
6277based on a parametric (probabilistic) model, e.g. based on Neyman-Pearson lemma, likelihood ratio, Receiver Operating Characteristic [ROC] curve plotting a False Acceptance Rate [FAR] versus a False Reject Rate [FRR]
G06K 9/629
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
9Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
62Methods or arrangements for recognition using electronic means
6288Fusion techniques, i.e. combining data from various sources, e.g. sensor fusion
629of extracted features
G06N 20/00
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
20Machine learning
Applicants
  • NORWEGIAN UNIVERSITY OF SCIENCE AND TECHNOLOGY (NTNU) [NO]/[NO]
Inventors
  • RAJA, Kiran Bylappa
  • RAMACHANDRA, Raghavendra
  • VENKATESH, Sushma
  • BUSCH, Christoph
Agents
  • JACKSON, Robert
Priority Data
1710560.230.06.2017GB
Publication Language English (en)
Filing Language English (EN)
Designated States
Title
(EN) DETECTION OF MANIPULATED IMAGES
(FR) DÉTECTION D'IMAGES MANIPULÉES
Abstract
(EN) An apparatus (30) for detecting morphed or averaged images, wherein the morphed or averaged images are synthetically generated images comprising information from two or more different source images corresponding to two or more subjects. The apparatus comprises: a feature extraction module for receiving an input image (33) and outputting a set of descriptor feature(s) characteristic of the image; and a classifier module (36) configured to allocate the input image either to a first class indicating that the image has been morphed or averaged or a second class indicating that it has not been morphed or averaged, based on the descriptor feature(s). The feature extraction module comprises a plurality of neural networks (31, 32) providing complementary descriptor feature(s) to the classifier module. The apparatus further comprises a fusion module (35) for combining descriptor feature data from each neural network and transmitting the fused feature data to the classifier module. The classifier module comprises a machine-learning system trained to classify the images using a training data set comprising morphed or averaged images and images that have not been morphed or averaged.
(FR) La présente invention concerne un appareil (30) pour détecter des images morphées ou moyennées, les images morphées ou moyennées étant des images générées de manière synthétique comprenant des informations provenant de deux ou plusieurs images sources différentes correspondant à au moins deux sujets. L'appareil comprend : un module d'extraction de caractéristiques pour recevoir une image d'entrée (33) et délivrer en sortie un ensemble de caractéristique(s) descriptive(s) caractéristique(s) de l'image; et un module de classification (36) configuré pour affecter l'image d'entrée soit à une première catégorie indiquant que l'image a été morphée ou moyennée, soit à une seconde catégorie indiquant qu'elle n'a pas été morphée ni moyennée, sur la base de la ou des caractéristiques descriptives. Le module d'extraction de caractéristiques comprend une pluralité de réseaux neuronaux (31, 32) fournissant une ou des caractéristiques descriptives complémentaires au module de classification. L'appareil comprend en outre un module de fusion (35) pour combiner des données associées aux caractéristiques descriptives en provenance de chaque réseau neuronal et transmettre les données associées aux caractéristiques fusionnées au module de classification. Le module de classification comprend un système d'apprentissage automatique entraîné pour classifier les images à l'aide d'un ensemble de données d'apprentissage comprenant des images morphées ou moyennées et des images qui n'ont pas été morphées ni moyennées.
Latest bibliographic data on file with the International Bureau