José L. Vásquez

First approach to automatic identification of bat calls, in the Basque Country

The conservation and preservation of biodiversity, its global and fair use, ensures the survival of the planet and ultimately ours. Taken into account the tools provided by scientific and technological developments and increasing concern for environmental conservation, it is of great relevance to identify appropriate technological tools for environmental sustainability. Moreover it is essential establishing certain protocols for their implementation and use. For this reason specialists in intelligent automatic systems, in biology and engineers have been working together for several years in this issue. This paper presents the first steps towards the development of a biometric system based on ultrasound for identification of bat calls in the Basque Country. They have been carried out three different kind of identification with regard to: specie, individual and behaviour.

Off line writer identification based on graphometric parameters

In this work is proposed a writer identification approach based on graphometrical and forensic features. The proposal replies to an off-line system, where the handwriting is provided before to perform the analysis. An Artificial Neural Network is used as classifier and after the decision fusion module, the system reaches up to 94.6% of success rate for a own database composed by 100 users with 10 samples per each one.

Writer identification on mobile device based on handwritten

This paper deals with exploring of the potential of writer identification by handwriting on a touch-screen phone for an application in access control systems. Our aim was to examine the possibility of writer recognition by a biometric model based on handwritten password. A mobile phone-server solution based on distributed blocks is proposed. The implemented approach performs a pre-processing block, in order to segment the handwritten password on the mobile phone. It also applies a feature extraction in order to have our biometric in-formation, running on the server. The classification is done with 10 online and offline features and is classified by a Naive Bayes classifier. We have used a database of 108 handwritten genuine (12 samples came from nine users) and 36 impostors (four false samples from nine users) written on a HTC Desire mobile phone with Android 2.2. The proposed system reached an accuracy of 96.87% in writer verification. The false acceptance rate of the proposed system is 11.11%.

Artificial neural network in FPGA for temperature prediction

In this work a temperature predictor has been designed. The prediction is made by an artificial neural network multilayer perceptron. Initially, the floating point algorithm was evaluated. Afterwards, the fixed point algorithm was designed on a Field Programmable Gate Array (FPGA). The architecture was fully parallelized and a maximum delay of 74 ns was obtained. The design tool used is System Generator of Xilinx.

Using calligraphies features for off line writer identification

This work proposes an off-line writer identification approach based on graphometrical and forensic features. We selected a set of features with independence of the text and some stability degree to natural changes in the writing. The system uses the LS-SVM classifier with RBF kernel, reaching up to 99.1% of success rate for an own database composed by 100 users with 10 samples per each one.

Writer identification approach by holistic graphometric features using off-line handwritten words

The biometric identification is an important topic with applications in different fields. Among the different modalities, based-handwriting biometric is a very useful and extended modality, and the most known one is the signature. The use of handwritten texts is researched presenting a biometric system for identifying writers from their handwritten words. A set of feature-based graphometric information has been extracted from off-line handwritten words to implement an automatic biometric approach. Given the handwritten nature of the information and its great variability, a feature selection based on principal component analysis and neural network classifier has been proposed. A fusion block based on neural networks has been added in order to reduce the effect of the data variability due to an increase and stabilization of the accuracy. A dataset composed of 100 writers have been used for the experiments. A holdout cross-validation was applied and the accuracy reached between 99.80% and 100%.

Applying forensic features on writer identification

This work present new parameters based on biometrie handwritten information for the writer identification. The feature extraction is developed by new algorithms based on image processing techniques. The handwritten parameters will be classified by artificial neural network and fusion strategy in order to increase the accuracy. After experiments, and using a dataset composed by 100 writers, this proposal reaches an accuracy of 82.7%.

Temperature Prediction Based on Different Meteorological Series

In this work, a temperature predictor has been designed and implemented based on different series of meteorological data. The prediction is built by an artificial neural network multilayer perceptron, using 5 samples as window size of meteorological data. Besides, the floating point algorithm was evaluated, reaching a mean square error of 0.35, meaning a variation of 0.28 Celsius degrees versus the real temperature. Different approaches will be applied in order to show our best proposal.

Design Methodology with System Generator in Simulink of a FHSS Transceiver on FPGA

This study aims to describe a design method for Field Programmable Gate Array (FPGA) (Maxfield, 2004) applied, in particular, to the design of a Frequency Hopping Spread Spectrum (FHSS) transceiver (Simon et al., 1994). Simulink (MathWorks, 2011) is a tool integrated in Matlab, which allows the design of systems using block diagrams in a fast and flexible way. Xilinx is one of the most important FPGA manufacturers and provides System Generator (Xilinx, 2011), it is a design environment over Simulink for FPGA based on the method described. The design is based on a previous FHSS transceiver designed for indoor wireless optical communications made with discrete components (Perez et al., 2003). One of the improvements in the proposed system is the physical integration.