Visa International Service Association

Konica Minolta Laboratory U.s.a., Inc.

Saman Sarraf

Kim R Wagner

Sunpreet Singh Arora

Lacey Best-Rowden

Duanduan Yang

A method and system of using a vehicle mounted camera device to authenticate a user during an interaction is disclosed. The method includes receiving interaction data regarding an interaction between a user operating a communication device and an access device, the user being near other candidate users. The method then includes determining one or more match indicators, the match indicators generated by comparing different sample biometric templates of the user with different enrolled biometric templates. At least one of the different biometric sample templates may be an image-based biometric template and at least one may be a voice print biometric template. Then the method includes identifying the user based on at least the match indicator associated with the voice print biometric template. The method then includes, if the match indicators are positive match indicators, initiating a process on behalf of the user.

Camera device enabled identification and disambiguation system and method

Described herein are a system and techniques for identifying and preventing certain fraud attacks that may be used to defeat facial recognition systems. In embodiments of the system described herein, biometric data may be segregated into regions, which are then processed separately and in parallel. Likeness scores are determined for each of the separate regions. By tracking individual region likeness scores used in access requests in accordance with embodiments of the disclosure, the system is able to identify potential fraud attacks that cannot be detected using conventional systems.

Defense mechanism against component-wise hill climbing using synthetic face generators

Defense mechanism against component wise hill climbing using synthetic face generators

In an intelligent character recognition (ICR) method for recognizing hand-written text images using a long-short term memory (LSTM) recurrent neural network (RNN), text images are segmented into text line images, and the text lines images are pre-processed to normalize the line height and to equalize the word spacings in each text line. Both training images used to train the RNN network and test images containing text to be recognized by the trained RNN network are pre-processed to have identical heights and identical word spacings between words. This method improves character recognition accuracy.

Text image processing using word spacing equalization for ICR system employing artificial neural network

Disclosed herein is a method of optimizing data normalization by selecting the best height normalization setting from training RNN (Recurrent Neural Network) with one or more datasets comprising multiple sample images of handwriting data, which comprises estimating a few top place ratios for normalization by minimizing a cost function for any given sample image in the training dataset, and further, determining the best ratio from the top place ratios by validating the recognition results of sample images with each top place ratio.

Growing community of inventors

Belmont, CA, United States of America

Saman Sarraf