J. Gorbe-Moya

TL;DR: The use of hybrid Hidden Markov Model (HMM)/Artificial Neural Network (ANN) models for recognizing unconstrained offline handwritten texts and new techniques to remove slope and slant from handwritten text and to normalize the size of text images with supervised learning methods are presented.

TL;DR: The current state of the UJIpenchars database, whose first version contains online representations of 1,364 isolated handwritten characters produced by 11 writers and is freely available at the UCI Machine Learning Repository, is described.

TL;DR: An efficient implementation of the Backpropagation (BP) algorithm to train Artificial Neural Networks with general feedforward topology is presented, which will lead to the "consecutive retrieval problem" that studies how to arrange efficiently sets into a sequence so that every set appears contiguously in the sequence.

TL;DR: A shoe is provided that provides an action whereby, when the wearer is running, the foot will strike the ground and the cleats will bend backward allowing the foot to continue its forward motion until it reaches a more gradual stop than normally occurs in any other running shoe.

TL;DR: The integration of multilayer perceptrons into the engine is presented, an improvement that speeds up the recognition process by taking advantage of the independence of these networks’ classification times from training set sizes.