Fundamenta Informaticae - Volume 101, issue 4

Authors: Konstantinidis, Stavros | Silva, Pedro V.

Article Type: Research Article

Abstract: A (combinatorial) channel consists of pairs of words representing all possible input-output channel situations. In a past paper, we formalized the intuitive concept of “largest amount of errors” detectable by a given language L, by defining the maximal error-detecting capabilities of L with respect to a given class of channels, and we showed how to compute all maximal error-detecting capabilities (channels) of a given regular language with respect to the class of rational channels and a class of channels involving only the substitution-error type. In this paper we resolve the problem for channels involving any combination of the basic error …types: substitution, insertion, deletion. Moreover, we consider the problem of finding the inverses of these channels, in view of the fact that L is error-detecting for γ if and only if it is error-detecting for the inverse of γ. We also discuss a natural method of reducing the problem of computing (inner) distances of a given regular language L to the problem of computing maximal error-detecting capabilities of L. Show more

Keywords: algorithm, automaton, combinatorial channel, edit string, error detection, maximal, regular language, string distance

DOI: 10.3233/FI-2010-287

Citation: Fundamenta Informaticae, vol. 101, no. 4, pp. 257-270, 2010

Authors: Kursa, Miron B. | Jankowski, Aleksander | Rudnicki, Witold R.

Article Type: Research Article

Abstract: Machine learning methods are often used to classify objects described by hundreds of attributes; in many applications of this kind a great fraction of attributes may be totally irrelevant to the classification problem. Even more, usually one cannot decide a priori which attributes are relevant. In this paper we present an improved version of the algorithm for identification of the full set of truly important variables in an information system. It is an extension of the random forest method which utilises the importance measure generated by the original algorithm. It compares, in the iterative fashion, the importances of original attributes …with importances of their randomised copies. We analyse performance of the algorithm on several examples of synthetic data, as well as on a biologically important problem, namely on identification of the sequence motifs that are important for aptameric activity of short RNA sequences. Show more

DOI: 10.3233/FI-2010-288

Citation: Fundamenta Informaticae, vol. 101, no. 4, pp. 271-285, 2010

Authors: Maulik, Ujjwal | Chakraborty, Debasis

Article Type: Research Article

Abstract: Satellite image classification is a complex process that may be affected by many factors. This article addresses the problem of pixel classification of satellite images by a robust multiple classifier system that combines k-NN, support vector machine (SVM) and incremental learning algorithm (IL). The effectiveness of this combination is investigated for satellite imagery which usually have overlapping class boundaries. These classifiers are initially designed using a small set of labeled points. Combination of these algorithms has been done based on majority voting rule. The effectiveness of the proposed technique is first demonstrated for a numeric remote sensing data described in …terms of feature vectors and then identifying different land cover regions in remote sensing imagery. Experimental results on numeric data as well as two remote sensing data show that employing combination of classifiers can effectively increase the accuracy label. Comparison is made with each of these single classifiers in terms of kappa value, accuracy, cluster quality indices and visual quality of the classified images. Show more

Keywords: Support vector machines, Incremental learning, Kernel function, Remote sensing imagery, Pixel classification, Multiple classifier system

DOI: 10.3233/FI-2010-289

Citation: Fundamenta Informaticae, vol. 101, no. 4, pp. 286-304, 2010

Authors: Shanker, N.R. | Ramakrishnan, S.S.

Article Type: Research Article

Abstract: The objective of the paper is to demonstrate how an image enhancement technique denoted quantum Fourier transform, improves the possibilities for visual interpretation of multispectral IKONOS satellite image, covering an urban area. Image representation is usually based on a mathematical function that has two spatial values, x and y. The spatial points x and y in the image function have the values to represent intensity of the image at that spatial point. An alternative method of image representation is the transformation. Transformation is also a mathematical method in which an image is represented as complex exponential summations. The complex exponentials …have the varying magnitudes, frequencies and phases. Transformation method of image representation is useful for image analysis, interpretations, filter etc. In this paper an image is represented based on the quantum mechanics for image analysis. The image is transformed using quantum Fourier transforms. For a specific application the result of processed image is better than the original image then it will be called as Enhancement. Enhancements of images are done in two broad categories. They are spatial domain methods and frequency domain methods. Frequency domain methods are based on transforms. And here quantum transforms is implemented. It is shown how the quantum Fourier transforms are applicable in the remote sensing satellite images, particularly for image enhancement. The satellite image of IKONOS is covering the Chennai city in Tamilnadu, India. From experimental result, it is observed that the pixels have low correlation when quantum Fourier transforms, are applied. Show more

DOI: 10.3233/FI-2010-290

Citation: Fundamenta Informaticae, vol. 101, no. 4, pp. 305-320, 2010

Authors: Ye, Guodong

Article Type: Research Article

Abstract: This paper presents a novel image encryption scheme based on Toeplitz matrix and Hankel matrix. According to the definition and the expression of Toeplitz matrix and Hankel matrix, we permutate the positions of image pixels to confuse the relationship between the plain-image and cipher-image, and construct a new key. To enhance security further, hyper-chaos system of Lü and Chen is taken to change the grey distribution of image pixels simultaneously. Numerical experimental results demonstrate that the key space is large enough, and the key is sensitive to initial conditions to resist the brute-force attack in the proposed scheme. Additionally, the …distribution of grey values in encrypted image has a random-like behavior to resist statistical analysis. The proposed encryption scheme in this paper can suggest a high secure and efficient way for real-time image encryption and transmission in practice. Show more

Keywords: image encryption scheme, Toeplitz matrix, Hankel matrix, Lü and Chen’s chaotic system, security analysis

DOI: 10.3233/FI-2010-291

Citation: Fundamenta Informaticae, vol. 101, no. 4, pp. 321-333, 2010