A feature extraction unsupervised neural network for an environmental data set

Acciani, G.; Chiarantoni, E.; Fornarelli, G.; Vergura, Silvano

doi:10.1016/s0893-6080(03)00014-5

Cited by 21 publications

(7 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…For this reason FUKUNAGA and KOONTZ (1970) proposed a data normalisation technique to use before PCA that may emphasise the differences between groups. The extension to composite populations may also take advantage of different techniques that have been proposed in the literature for dimension reduction, such as neural networks (CASTELLANO and FANELLI, 2000;ACCIANI et al, 2003), second-order statistical methods (FODOR, 2002), higher-order statistical methods (independent component analysis, projection pursuit) (HYVARINEN, 1999) and methods based on wavelet transforms (COCCHI et al, 2003).…”

Section: Future Developmentsmentioning

confidence: 99%

Feature selection of stabilometric parameters based on principal component analysis

Rocchi

Chiari

Cappello

2004

Med. Biol. Eng. Comput.

193

132

View full text Add to dashboard Cite

This study addresses the challenge of identifying the features of the Centre of pressure (COP) trajectory that are most sensitive to postural performance, with the aim of avoiding redundancy and allowing a straightforward interpretation of the results. Postural sway in 50 young, healthy subjects was measured by a force platform. Thirty-seven stabilometric parameters were computed from the one-dimensional and two-dimensional COP time series. After normalisation to the relevant biomechanical factors, by means of multiple regression models, a feature selection process was performed based on principal component analysis. Results suggest that COP two-dimensional time series can be primarily characterised by four parameters, describing the size of the COP path over the support surface; the principal sway direction; and the shape and bandwidth of the power spectral density plot. COP one-dimensional time series (antero-posterior (AP) and medio-lateral (ML)) can be characterised by six parameters describing COP dispersion along the AP direction; mean velocity along the ML and AP directions; the contrast between ML and AP regulatory activity; and two parameters describing the spectral characteristics of the COP along the AP direction. On the basis of the results obtained, some guidelines are suggested for the choice of stabilometric parameters to use, with the aim of promoting standardisation in quantitative posturography.

show abstract

Section: Future Developmentsmentioning

confidence: 99%

Feature selection of stabilometric parameters based on principal component analysis

Rocchi

Chiari

Cappello

2004

Med. Biol. Eng. Comput.

193

132

View full text Add to dashboard Cite

show abstract

“…Una gran cantidad de estos trabajos sobresalen por el uso de técnicas no supervisadas para resolver el problema; empleando algoritmos tales como K-Means [8], ISODATA [9], fuzzy-K-Means [10] y redes neuronales auto-organizativas SOM [11], [12], [2], [13], [6]. Es importante resaltar que no existe una definición estándar de tipos de relieve, por lo tanto, cada clasificación lleva consigo un componente de subjetividad inherente a la geografía del caso de estudio para el que fue desarrollado.…”

Section: Clasificación De Tipos De Relieveunclassified

“…Además de esta información, también varias investigaciones en donde se estudia la generación de nuevos descriptores que tal que estos ofrezcan mejores resultados al momento de realizar la clasificación. Para dar algunos ejemplos se pueden relacionar los trabajos de Acciani et al [11] y Ardiansyah et al [15] y Renno et al [16].…”

Section: Clasificación De Tipos De Relieveunclassified

Automatic landform clasification of uplands based on Haralick's texture

Patiño

Mery

Botero-Fernández

et al. 2012

2012 XXXVIII Conferencia Latinoamericana en Informatica (CLEI)

View full text Add to dashboard Cite

This paper presents the results of applying Haralik's textures in the upland landform classification process. In environmental and landuse projects is important to know what kind of landforms are present in a geographic area. The objective of this work is to improve the performance of such classifications, adding information to the commonly used data features in such problems. The texture information was extracted using the method of Haralick, parameterized with moving windows of geo-referenced maps in raster format. Those maps contains information of the terrain morphology such as elevation, slopes, etc. Several tests were performed using different classifiers and cross-validation over a dataset with a total of 203401 samples. It was shown that Haralick's textures features are useful for the problem, because its performance is higher (about 97 %) in comparison with the achieved when using only morphological information of the terrain. I. INTRODUCCIÓNLa clasificación automática de tipos de relieve surge de la necesidad de conocer la forma de un territorio evitando recorrerlo por completo; ya sea por ser de gran tamaño, que requiera un esfuerzo considerable o que sea de difícil o imposible acceso [1], [2]. Este tipo de clasificaciones consitituyen un insumo importante para procesos de toma de decisiones eń areas ambientales o de ordenamiento territorial; además, sirven como información base para diversos modelos hidrológicos y geológicos [3], [4].En este trabajo se presenta un enfoque de clasificación de tipos de relieve, específicamente aquellos que corresponden a zonas de alta, media y baja montaña. Las dos clases objetivo del problema de clasificación plateado son: cañones (valles de quebradas y ríos jóvenes) y vertientes (Laderas de montañas que contribuyen al cauce de los ríos y quebradas).El reconocimiento de patrones es el proceso a través del cual se construye un modelo que aproxima la forma en que los individuos de un conjunto de datos son asignados a clases (patrones) previamente definidas o desconocidas. La clasificación de patrones es la etapa final del proceso de reconocimiento y consiste en asignar una etiqueta a un vector n-dimensional donde cada dimensión representa una característica (descriptor o feature) de la información que se está analizando, y cada etiqueta representa una de las clases.Esta investigación propone un mecanismo para encontrar un conjunto de descriptores que ofrezcan un considerable grado de separabilidad de las clases objetivo haciendo uso de información sobre la textura del terreno. Una textura de terreno puede ser definida como el patrón de comportamiento que se percibe de la forma de una región geográfica.Esta información es calculada a partir del análisis de diferentes mapas georeferenciados de variables geomorfológicas en formato raster; e.g: MDE (Modelo Digital de Elevación), pendientes, curvaturas, etc. El conjunto de datos para la clasificación propuesta se construye entonces como la unión entre las características geomorfológicas y las de características de textura der...

show abstract

“…Since it has been demonstrated that the ozone level above some well known threshold is rather harmful for human health, the analysis of environmental data has become an emergent issue for many years [1][2][3][4]. It is devoted to the determination of an elevated level of ground ozone (O3) which represents a potential danger not only to human health, but also to crops and vegetation.…”

Section: Introductionmentioning

confidence: 99%

Feature extraction for atmospheric pollution detection

Ferchichi¹,

Zidi

Laabidi³

et al. 2011

2011 International Conference on Communications, Computing and Control Applications (CCCA)

View full text Add to dashboard Cite

Atmospheric data sets are represented by an amount of heterogeneous and redundant data. As number of measurements grows, a strategy is needed to select and efficiently analyze the useful information from the whole data set. The aim of this work is to propose a feature extraction technique based on construction of clusters of similar features. The main objective of the proposed process is to attempt to reach a more accurate classification task and to achieve a more compact representation of the underlying structure of the data. The paper reports the results obtained using the above extraction and analysis procedure of a real data set on atmospheric pollution. It is shown that the proposed approach is able to detect underlying relationship between features and thus get to ameliorate classification accuracy rate.

show abstract

A feature extraction unsupervised neural network for an environmental data set

Cited by 21 publications

References 26 publications

Feature selection of stabilometric parameters based on principal component analysis

Feature selection of stabilometric parameters based on principal component analysis

Automatic landform clasification of uplands based on Haralick's texture

Feature extraction for atmospheric pollution detection

Contact Info

Product

Resources

About