Proposed Classification Procedure Research Articles

Exploiting Spectral and Spatial Information in Hyperspectral Urban Data With High Resolution

Very high resolution hyperspectral data should be very useful to provide detailed maps of urban land cover. In order to provide such maps, both accurate and precise classification tools need, however, to be developed. In this letter, new methods for classification of hyperspectral remote sensing data are investigated, with the primary focus on multiple classifications and spatial analysis to improve mapping accuracy in urban areas. In particular, we compare spatial reclassification and mathematical morphology approaches. We show results for classification of DAIS data over the town of Pavia, in northern Italy. Classification maps of two test areas are given, and the overall and individual class accuracies are analyzed with respect to the parameters of the proposed classification procedures.

Company Characteristics and Common Stock Returns: The Indian Experience

The paper examines the relationship between selected company characteristics and common stock returns. The empirical results suggest that there is a strong size effect in the Indian stock market using both market-based as well as non-market based measures of company size. We also detect a weak value effect on stock returns, especially when E/P ratio is employed as a relative distress proxy. The study further finds that the present stock classification system in India fails to differentiate in returns on different categories of stocks. We recommend an alternative stock classification system based on company size and relative distress. The proposed classification procedure will provide better insights to investors about the risk-return characteristics of common stocks.

Robust Rotation-Invariant Texture Classification Using a Model Based Approach

In this paper, a model based texture classification procedure is presented. The texture is modeled as the output of a linear system driven by a binary image. This latter retains the morphological characteristics of the texture and it is specified by its spatial autocorrelation function (ACF). We show that features extracted from the ACF of the binary excitation suffice to represent the texture for classification purposes. Specifically, we employ a moment invariants based technique to classify the ACF. The resulting proposed classification procedure is thus inherently rotation invariant. Moreover, it is robust with respect to additive noise. Experimental results show that this approach allows obtaining high correct rotation-invariant classification rates while containing the size of the feature space.

Clustering financial time series: an application to mutual funds style analysis

Classification can be useful in giving a synthetic and informative description of contexts characterized by high degrees of complexity. Different approaches could be adopted to tackle the classification problem: statistical tools may contribute to increase the degree of confidence in the classification scheme. A classification algorithm for mutual funds style analysis is proposed, which combines different statistical techniques and exploits information readily available at low cost. Objective, representative, consistent and empirically testable classification schemes are strongly sought for in this field in order to give reliable information to investors and fund managers who are interested in evaluating and comparing different financial products. Institutional classification schemes, when available, do not always provide consistent and representative peer groups of funds. A “return-based” classification scheme is proposed, which aims at identifying mutual funds’ styles by analysing time series of past returns. The proposed classification procedure consists of three basic steps: (a) a dimensionality reduction step based on principal component analysis, (b) a clustering step that exploits a robust evolutionary clustering methodology, and (c) a style identification step via a constrained regression model first proposed by William Sharpe. The algorithm is tested on a sample of Italian mutual funds and achieves satisfactory results with respect to (i) the agreement with the existing institutional classification and (ii) the explanatory power of out of sample variability in the cross-section of returns.

Elastodynamic inversion for shape reconstruction and type classification of flaws

Two linearized inverse scattering methods are investigated to reconstruct the shape of flaws in the elastic solid. One is based on the Born approximation and the other is based on the Kirchhoff approximation. The Born inversion is sensitive to the volumetric flaw but not to a crack-like flaw. On the other hand, the Kirchhoff inversion reacts to both boundaries of volumetric and crack-like flaws. The combined use of Born and Kirchhoff inversions leads to the classification method of flaw type. The performance of the proposed classification procedure is demonstrated by the numerical simulations and then by the experimental measurements for the two-dimensional models of flaws. An example for the three-dimensional shape reconstruction is also shown by using the numerically calculated backscattering amplitudes.

A Functional Approach to Soil Characterization in Support of Precision Agriculture

Managing soil variability is an integral aspect of precision agriculture (PA). Existing soil databases, however, are found to match few of the requirements for PA. The nature of these requirements and their implications for soil information need to be further explored. Ongoing developments towards a decision support system (DSS) for PA in the Netherlands have shed some light on this issue. Two soil related DSS‐components are presented: (i) the construction of a soil database at the farm level and (ii) the delineation of soil functional units at the field level. Developed methods were tested in a case study for two arable fields located on Dutch marine clay soils. Basic soil data were collected in a 1:5000 soil survey and supplemented with secondary data derived through pedotransfer functions. Soil characterization focused on functional properties describing soil‐specific characteristics in terms of water regimes and nutrient dynamics. Four properties were considered: (i) water stress, (ii) N‐stress, (iii) N‐leaching and (iv) residual N‐content at harvest. These were quantified for individual soil profiles using a mechanistic–deterministic simulation model. Sensitivity to water stress was evaluated for a dry year (1989), other properties were quantified for a wet year (1987). Based on functional similarity, the soil profiles were grouped into functional classes using a fuzzy c‐means classifier. Standard interpolation techniques and a boundary detection algorithm subsequently identified soil functional units in each field. Analysis of variance revealed that >65% of the spatial variation could thus be accounted for. This confirmed that (i) the proposed classification procedure was efficient and (ii) soil functional units are suitable entities to be used as management units for PA.

A bootstrap generalized likelihood ratio test in discriminant analysis

A generalized likelihood ratio test is developed for classification in two populations when one needs to control one of the probabilities of misclassification. The proposed classification procedure is constructed by applying the parametric bootstrap to the generalized likelihood ratio. There are known methods for controlling this misclassification probability for the case where normal distributions with the same covariance matrix are assumed. Our approach, however, can be applied to not only this case but to the case of normal distributions with different covariance matrices and the case of a mixture of discrete and continuous variables. The results given here do not depend on normality but can, in fact, be applied to any distribution for which the maximum likelihood estimates exist. We do, however, restrict our simulation of these results to the normal distribution if the variates are all continuous. Three cases are simulated: normal distributions with equal covariance matrix, normal distributions with unequal covariance matrices, and mixture of categorical and normal variables. An application to classifying seismic events is presented.