Jm Jose Palomares

Improving the accuracy while preserving the interpretability of fuzzy function approximators by means of multi-objective evolutionary algorithms

The identification of a model is one of the key issues in the field of fuzzy system modeling and function approximation theory. An important characteristic that distinguishes fuzzy systems from other techniques in this area is their transparency and interpretability. Especially in the construction of a fuzzy system from a set of given training examples, little attention has been paid to the analysis of the trade-off between complexity and accuracy maintaining the interpretability of the final fuzzy system. In this paper a multi-objective evolutionary approach is proposed to determine a Pareto-optimum set of fuzzy systems with different compromises between their accuracy and complexity. In particular, two fundamental and competing objectives concerning fuzzy system modeling are addressed: fuzzy rule parameter optimization and the identification of system structure (i.e. the number of membership functions and fuzzy rules), taking always in mind the transparency of the obtained system. Another key aspect of the algorithm presented in this work is the use of some new expert evolutionary operators, specifically designed for the problem of fuzzy function approximation, that try to avoid the generation of worse solutions in order to accelerate the convergence of the algorithm.

A novel method to determine the electron temperature and density from the absolute intensity of line and continuum emission: application to atmospheric microwave induced Ar plasmas

An absolute intensity measurement (AIM) technique is presented that combines the absolute measurements of the line and the continuum emitted by strongly ionizing argon plasmas. AIM is an iterative combination of the absolute line intensity–collisional radiative model (ALI–CRM) and the absolute continuum intensity (ACI) method. The basis of ALI–CRM is that the excitation temperature T13 determined by the method of ALI is transformed into the electron temperature Te using a CRM. This gives Te as a weak function of electron density ne. The ACI method is based on the absolute value of the continuum radiation and determines the electron density in a way that depends on Te. The iterative combination gives ne and Te. As a case study the AIM method is applied to plasmas created by torche a injection axiale (TIA) at atmospheric pressure and fixed frequency at 2.45 GHz. The standard operating settings are a gas flow of 1 slm and a power of 800 W; the measurements have been performed at a position of 1 mm above the nozzle. With AIM we found an electron temperature of 1.2 eV and electron density values around 1021 m−3. There is not much dependence of these values on the plasma control parameters (power and gas flow). From the error analysis we can conclude that the determination of Te is within 7% and thus rather accurate but comparison with other studies shows strong deviations. The ne determination comes with an error of 40% but is in reasonable agreement with other experimental results.

Atmospheric microwave-induced plasmas in Ar/H2 mixtures studied with a combination of passive and active spectroscopic methods

Several active and passive diagnostic methods have been used to study atmospheric microwave-induced plasmas created by a surfatron operating at a frequency of 2.45 GHz and with power values between 57 and 88 W. By comparing the results with each other, insight is obtained into essential plasma quantities, their radial distributions and the reliability of the diagnostic methods. Two laser techniques have been used, namely Thomson scattering for the determination of the electron density, ne, and temperature, Te, and Rayleigh scattering for the determination of the heavy particle temperature, Tg. In combination, three passive spectroscopic techniques are applied, the line broadening of the Hβ line to determine ne, and two methods of absolute intensity measurements to obtain ne and Te. The active techniques provide spatial resolution in small plasmas with sizes in the order of 0.5 mm. The results of ne measured with three different methods show good agreement, independent of the plasma settings. The Te values obtained with two techniques are in good agreement for the condition of a pure argon plasma, but they show deviations when H2 is introduced. The introduction of a small amount (0.3%) of H2 into an argon plasma induces contraction, reduces ne, increases Te, enhances the departure from equilibrium and leads to conditions that are close to those found in cool atmospheric plasmas.

Global motion estimation algorithm for video segmentation

This work presents a new global motion estimation algorithm for MPEG compressed video sequences. It makes use of a feature extraction technique based on the Generalized Hough Transform, which is able to provide rotation, scale and displacement parameters when comparing two different frames from a video sequence. Thus, pan, tilt, swing (rotation along z-axis) and zoom effects will be studied using the proposed algorithm. DC coefficients of the DCT transform are extracted from the MPEG stream and used to create DC images, which are the starting point for the global motion estimation algorithm. Further application of the feature extraction technique to DC images will allow to perform motion estimation, reducing processing time as the decompression process is avoided. Pseudocode and details about the implementation of the algorithm as well as statistics to illustrate the efficiency of the algorithm are provided.

Temperature-Compensated Clock Skew Adjustment

This work analyzes several drift compensation mechanisms in wireless sensor networks (WSN). Temperature is an environmental factor that greatly affects oscillators shipped in every WSN mote. This behavior creates the need of improving drift compensation mechanisms in synchronization protocols. Using the Flooding Time Synchronization Protocol (FTSP), this work demonstrates that crystal oscillators are affected by temperature variations. Thus, the influence of temperature provokes a low performance of FTSP in changing conditions of temperature. This article proposes an innovative correction factor that minimizes the impact of temperature in the clock skew. By means of this factor, two new mechanisms are proposed in this paper: the Adjusted Temperature (AT) and the Advanced Adjusted Temperature (A2T). These mechanisms have been combined with FTSP to produce AT-FTSP and A2T-FTSP Both have been tested in a network of TelosB motes running TinyOS. Results show that both AT-FTSP and A2T-FTSP improve the average synchronization errors compared to FTSP and other temperature-compensated protocols (Environment-Aware Clock Skew Estimation and Synchronization for WSN (EACS) and Temperature Compensated Time Synchronization (TCTS)).

Thomson scattering measurements on a low pressure surface wave sustained plasma in argon

Thomson scattering (TS) experiments have been made on a low pressure surfatron induced plasma. TS is an active diagnostic method and the experimental results are directly related to important plasma properties such as the electron density, ne, and the electron temperature, Te. Therefore, the TS results for ne and Te can be used to calibrate passive diagnostic methods which are often based on plasma models. However, to apply TS on a surfatron induced plasma inside a quartz tube is experimentally demanding because of the large amount of stray light and a low intensity of the TS signal. To achieve low detection limits and high stray light rejection, a triple grating spectrograph was used in the detection branch and an iCCD was used to record the TS spectrum. For a typical plasma condition with an argon pressure of 10 mbar and an absorbed power of 50 W, the measured electron density was found to be equal to ne ≈ 4 × 1019 m−3 and the electron temperature Te ≈ 1.2 eV. In addition, frame-averaged results for 6, 10, 15 and 20 mbar argon plasmas for absorbed microwave powers in between 25 ≤ Pab ≤ 60 W are presented. The trends found in the dependence of the pressure and power density are according to theory.

Parallelizing and optimizing LIP-canny using NVIDIA CUDA

The Canny algorithm is a well known edge detector that is widely used in the previous processing stages in several algorithms related to computer vision. An alternative, the LIP-Canny algorithm, is based on a robust mathematical model closer to the human vision system, obtaining better results in terms of edge detection. In this work we describe LIP-Canny algorithm under the perspective from its parallelization and optimization by using the NVIDIA CUDA framework. Furthermore, we present comparative results between an implementation of this algorithm using NVIDIA CUDA and the analogue using a C/C++ approach.

Skin lesion diagnosis using fluorescence images

This paper presents a computer aided diagnosis system for skin lesions. Diverse parameters or features extracted from fluorescence images are evaluated for cancer diagnosis. The selection of parameters has a significant effect on the cost and accuracy of an automated classifier. The genetic algorithm (GA) performs parameters selection using the classifier of the K-nearest neighbours (KNN). We evaluate the classification performance of each subset of parameters selected by the genetic algorithm. This classification approach is modular and enables easy inclusion and exclusion of parameters. This facilitates the evaluation of their significance related to the skin cancer diagnosis. We have implemented this parameter evaluation scheme adopting a strategy that automatically optimizes the K-nearest neighbours classifier and indicates which features are more relevant for the diagnosis problem.

The Use of Video-Gaming Devices as a Motivation for Learning Embedded Systems Programming

As embedded systems are becoming prevalent in everyday life, many universities are incorporating embedded systems-related courses in their undergraduate curricula. However, it is not easy to motivate students in such courses since they conceive of embedded systems as bizarre computing elements, different from the personal computers with which they are familiar. This problem has been overcome at the University of Granada, Spain, by taking advantage of the connection many students have with video games.

Polydiagnostic calibration performed on a low pressure surface wave sustained argon plasma

The electron density and electron temperature of a low pressure surface wave sustained argon plasma have been determined using passive and active (laser) spectroscopic methods simultaneously. In this way the validity of the various techniques is established while the plasma properties are determined more precisely. The electron density, ne, is determined with Thomson scattering (TS), absolute continuum measurements, Stark broadening and an extrapolation of the atomic state distribution function (ASDF). The electron temperature, Te, is obtained using TS and absolute line intensity (ALI) measurements combined with a collisional–radiative (CR) model for argon. At an argon pressure of 15 mbar, the ne values obtained with TS and Stark broadening agree with each other within the error bars and are equal to (4 ± 0.5) × 1019 m−3, whereas the ne value (2 ± 0.5) × 1019 m−3 obtained from the continuum is about 30% lower. This suggests that the used formula and cross-section values for the continuum method have to be reconsidered. The electron density determined by means of extrapolation of the ASDF to the continuum is too high (~1020 m−3). This is most probably related to the fact that the plasma is strongly ionizing so that the extrapolation method is not justified. At 15 mbar, the Te values obtained with TS are equal to 13 400 ± 1100 K while the ALI/CR-model yields an electron temperature that is about 10% lower. It can be concluded that the passive results are in good or fair agreement with the active results. Therefore, the calibrated passive methods can be applied to other plasmas in a similar regime for which active diagnostic techniques cannot be used.