Ncr Corporation

Chao He

Gary A Ross

Ping Chen

Mark Girolami

Andrew Colley

Alan R Greig

Meng Yu

Kristoffer Dominic Amora

Em Parac

Theodoros Damoulas

Richard C Quipanes

Wenjia Yin

Approaches to produce verification are addressed which may be suitably employed in conjunction with a low resolution image scanner. Features, such as lines, corners, text, texture, edges and the like are extracted and compared with features for an identified item or evaluated to determine if the features collectively are indicative of a produce item or a non-produce item.

Produce and non-produce verification using hybrid scanner

The present invention provides a method of validating a media item, the method comprising receiving image data corresponding to a plurality of digital images of a media item to be validated, wherein each image is associated with a respective one of a plurality of different channels of an illumination source for illuminating the media item, selecting image data corresponding to two of the plurality of images, dividing a common parameter of the selected image data associated with corresponding locations of the two selected images to create ratio image data representing a ratio image for two different channels of the illumination source, extracting feature information associated with the ratio image, and classifying the extracted feature information with at least one classifier of a validation template to determine whether or not the media item is valid. A computer program and a Self-Service Terminal for validating a media item, such as a banknote, are also provided.

Media item validation

A method and apparatus for identifying a media item. The apparatus includes a module for receiving media items and capturing an image of a received media item. The method involves ascertaining external dimensions of the media item from the captured image; creating a monochromatic image based on high contrast boundaries within the captured image; matching the monochromatic image to one of a plurality of templates of media items of approximately the same external dimensions as the media item; and providing details of the matched template, including type and orientation of the media item corresponding to the matched template.

Media identification

A produce recognition method which uses hierarchical Bayesian learning and kernel combination, and which offers classification-oriented synergistic data integration from diverse sources. An example method includes providing a classifier having a plurality of inputs, each input being adapted to receive produce data of a different modality; mapping the produce data to the respective input of the classifier by a computer; for each input, independently operating on the data relating to that input to create a feature set by the computer; comparing each feature in the feature set to respective pre-trained data for that feature to produce a similarity description set; combining all similarity description sets using a dedicated weighting function to produce a composite similarity description by the computer; and deriving a plurality of class values from the composite similarity description to create a recognition result for the produce item by the computer.

Produce recognition method

A method of validating a media item, comprising: for each of a plurality of sensors that sense a media item, determining a respective validation score indicating a likelihood that the sensed media item is valid; and determining if the media item is valid responsive to the plurality of validation scores. Apparatus for validating a media item and a document processing module are also provided.

A system and method of managing casino patron money balances which stores patron money balances in electronic form. The system includes an interrogator, and a self-service computer coupled to the interrogator for interrogating a wireless communicator associated with a casino patron to determine a money balance associated with the wireless communicator and for processing a selection by the casino patron to produce a change in the money balance.

System and method of managing casino patron money balances

A method of authenticating a media item is described. The method comprises: measuring a magnetic signal along the media item, converting the measured signal to a calibrated length signal, aligning the calibrated length signal with a reference signal, extracting features from the aligned signal, and classifying the media item based on the extracted features.

Media authentication

A method of categorizing defects in a media item. The method comprising the steps of: receiving a binarised image of the media item, where the binarised image comprises a plurality of pixels, each pixel having either a potential defect intensity or a non-defect intensity; and identifying one or more blobs comprising contiguous pixels each having a potential defect intensity. For each identified blob, the method involves comparing a size of the blob with a damage threshold; ignoring the blob if the blob size is smaller than the damage threshold; and for each identified blob having a size exceeding or equaling the damage threshold, categorizing the identified blob.

System and method of categorising defects in a media item

A method of detecting staining on a media item is described. An example method includes receiving an image of the media item, including a plurality of pixels having different intensity values within a range of intensity values, applying central weighting to the received image, applying a threshold to each pixel in the centrally-weighted image to transform each pixel to a binary value, comparing a pixel in the evaluation image with a pixel in a binary reference image to create a difference image including (i) a stain pixel at each spatial location in which a pixel in the evaluation image has a low intensity pixel and the corresponding pixel in the binary reference image has a high intensity pixel, and (ii) a non-stain pixel at all other spatial locations. The media item is identified as stained in the event that the difference image meets a staining criterion.

Stain detection

A technique for use in automatic validation of a media item involves accessing a template that comprises multiple one-class classifiers, each corresponding to one of multiple classes to which the media item might belong, and then applying each of the one-class classifiers to an image of the media item to generate a result set for each of the multiple classes. The result set for each media class is then analyzed to assess whether the media item belongs to that class.

Validation template for valuable media of multiple classes

A technique for use in automated recognition of a media item involves accessing a template that includes multiple segmentation maps that each is associated with one of multiple classes to which the media item might belong. For each of the multiple classes, the segmentation map is applied to an image of the media item to extract a feature set for the image, the feature set is analyzed, and an assessment is made as to whether the media item belongs to the class.

Method of recognizing a media item

A method and apparatus for evaluating a soiling level of a media item, such as a banknote. The method comprises receiving an image of a media item including a plurality of pixels, each pixel having an associated intensity. Two sets of pixels are created, where each pixel in the first set has a higher intensity than each pixel in the second set. A representative intensity value from the first set and a representative intensity value from the second set are calculated. A soiling value from the ratio of the representative intensity value of the second set to the representative intensity value of the first set is calculated. The media item is captured if the soiling value fulfils a capture criterion.

Evaluating soiling of a media item

Method of validating a media item

A method of creating a classifier for media validation is described. Information from all of a set of training images from genuine media items is used to form a segmentation map which is then used to segment each of the training set images. Features are extracted from the segments and used to form a classifier which is preferably a one-class statistical classifier. Classifiers can be quickly and simply formed, for example when the media is a banknote for different currencies and denominations in this way and without the need for examples of counterfeit banknotes. A media validator using such a classifier is described as well as a method of validating a banknote using such a classifier. In a preferred embodiment a plurality of segmentation maps are formed, having different numbers of segments. If higher quality counterfeit media items come into the population of media items, the media validator is able to react immediately by switching to using a segmentation map having a higher number of segments without the need for re-training.

Growing community of inventors

Dundee, United Kingdom

Chao He