Empirical model generation techniques for planar microwave components using electromagnetic linear regression models

Accurate and efficient empirical model generation techniques of microwave devices, for a large range of geometric and material parameters opportunely chosen, are presented. The empirical models are based on multiple linear regression approach, which compensates the error between an initial inaccurate empirical model and an electromagnetic (EM) full-wave solver (or measurement data). The aim of these techniques is to generate accurate empirical models, which are computationally very efficient with respect to any EM technique. These simple models could be integrated in a toolbox of any commercially available computed-aided design tools for RF/microwave circuits. Comparisons with artificial neural networks and linear-regression-based models are listed and discussed for the dispersion of a microstrip transmission line propagating the quasi-TEM mode and a microwave tunable phase shifter propagating the even mode.     

Supervised classification of marble textures using support vector machines

A novel approach is described for the supervised classification of marble textures in different classes according to visual appearance, using sum and difference histograms for texture analysis and feature extraction, and support vector machines for classification. Results show very good discrimination between classes.     

Wavelet and wavelet packet compression of phonocardiograms

The first reported method intended for the compression of the phonocardiogram, using wavelet coefficients thresholding, together with Huffman and run length encoding coding techniques to obtain further compression is described. Results show better compression rates and lower distortion than standard audio compression techniques.     

Classification of marble surfaces using wavelets

A new methodology for the classification of images of marble surfaces is described according to visual quality parameters based on multiresolution decomposition with wavelets. Experimental results show that the three marble categories considered can be easily separated with the proposed method.     

Automatic system for quality-based classification of marble textures

In this paper, we present an automatic system and algorithms for the classification of marble slabs into different groups in real time in production line, according to slabs quality. The application of the system is aimed at the marble industry, in order to automate and improve the manual classification process of marble slabs carried out at present. The system consists of a mechatronic prototype, which houses all the required physical components for the acquisition of marble slabs images in suitable light conditions, and computational algorithms, which are used to analyze the color texture of the marble surfaces and classify them into their corresponding group. In order to evaluate the color representation influence on the image analysis, four color spaces have been tested: RGB, XYZ, YIQ, and K-L. After the texture analysis performed with the sum and difference histograms algorithm, a feature extraction process has been implemented with principal component analysis. Finally, a multilayer perceptron neural network trained with the backpropagation algorithm with adaptive learning rate is used to classify the marble slabs in three categories, according to their quality. The results (successful classification rate of 98.9%) show very high performance compared with the traditional (manual) system.     

Encoding technique for binary sequences using vector tree partitioning applied to compression of phonocardiographic signals

Described is an efficient and low complexity technique to encode binary sequences like the significance map generated in wavelet-based compression methods. This is achieved using a generalised form of the bintree partitioning for 1-D signals, the vector K-tree partitioning. Results show higher compression rate and lower compression time than other binary sequences compression methods. Application to the compression of phonocardiograms for teletransmission also showed higher compression rate than set partitioning in hierarchical trees.