The search session has expired. Please query the service again.
Displaying 61 –
80 of
149
In this paper, we describe an efficient method for 3D image segmentation. The method uses a PDE model – the so called generalized subjective surface equation which is an equation of advection-diffusion type. The main goal is to develop an efficient and stable numerical method for solving this problem. The numerical solution is based on semi-implicit time discretization and flux-based level set finite volume space discretization. The space discretization is discussed in details and we introduce three...
Fuzzy clustering extends crisp clustering in the sense that objects can belong to various clusters with different membership degrees at the same time, whereas crisp or deterministic clustering assigns each object to a unique cluster. The standard approach to fuzzy clustering introduces the so-called fuzzifier which controls how much clusters may overlap. In this paper we illustrate, how this fuzzifier can help to reduce the number of undesired local minima of the objective function that is associated...
Fuzzy classification systems differ from fuzzy controllers in the form of their outputs. For classification problems a decision between a finite number of discrete classes has to be made, whereas in fuzzy control the output domain is usually continuous, i.e. a real interval. In this paper we consider fuzzy classification systems using the max-min inference scheme and classifying an unknown datum on the basis of maximum matching, i.e. assigning it to the class appearing in the consequent of the rule...
Fuzzy set theory, a recent generalization of classical set theory, has attracted the attention of researchers working in various areas including pattern recognition, which has had a seminal influence in the development of this new theory. This paper attempts to discuss some of the methodologies that have been suggested for pattern recognition, and techniques for image processing and speech recognition.
Classifiers can be combined to reduce classification errors. We did experiments on a data set consisting of different sets of features of handwritten digits. Different types of classifiers were trained on these feature sets. The performances of these classifiers and combination rules were tested. The best results were acquired with the mean, median and product combination rules. The product was best for combining linear classifiers, the median for -NN classifiers. Training a classifier on all features...
This work has been partially supported by Grant No. DO 02-275, 16.12.2008, Bulgarian
NSF, Ministry of Education and Science.Hausdorff distance (HD) seems the most efficient instrument
for measuring how far two compact non-empty subsets of a metric space are
from each other. This paper considers the possibilities provided by HD and
some of its modifications used recently by many authors for resemblance
between binary text images. Summarizing part of the existing word image
matching methods, relied...
The purpose of feature selection in machine learning is at least two-fold - saving measurement acquisition costs and reducing the negative effects of the curse of dimensionality with the aim to improve the accuracy of the models and the classification rate of classifiers with respect to previously unknown data. Yet it has been shown recently that the process of feature selection itself can be negatively affected by the very same curse of dimensionality - feature selection methods may easily over-fit...
Morphological Associative Memories have been proposed for some image denoising applications. They can be applied to other less restricted domains, like image retrieval and hyperspectral image unsupervised segmentation. In this paper we present these applications. In both cases the key idea is that Autoassociative Morphological Memories selective sensitivity to erosive and dilative noise can be applied to detect the morphological independence between patterns. Linear unmixing based on the sets of...
Segmentation is the art of partitioning an image into different regions where each one has some degree of uniformity in its feature space. A number of methods have been proposed and blind segmentation is one of them. It uses intrinsic image features, such as pixel intensity, color components and texture. However, some virtues, like poor contrast, noise and occlusion, can weaken the procedure. To overcome them, prior knowledge of the object of interest has to be incorporated in a top-down procedure...
Random forest is an ensemble method of machine learning that reaches a high level of accuracy in decision-making but is difficult to understand from the point of view of interpreting local or global decisions. In the article, we use this method as a means to analyze the edge 3-colorability of cubic graphs and to find the properties of the graphs that affect it most strongly. The main contributions of the presented research are four original datasets suitable for machine learning methods, a random...
The paper gives a new interpretation and a possible optimization of the well-known -means algorithm for searching for a locally optimal partition of the set which consists of disjoint nonempty subsets , . For this purpose, a new divided -means algorithm was constructed as a limit case of the known smoothed -means algorithm. It is shown that the algorithm constructed in this way coincides with the -means algorithm if during the iterative procedure no data points appear in the Voronoi diagram....
The present paper presents and discusses a methodology for interpreting the results, obtained from the application of a pattern classifier to an independent test set. It addresses the problem of testing the random classification null hypothesis in the multiclass case, by introducing an exact probability technique. The discussion of this technique includes the presentation of an interval estimation technique for the probability of correct classification, which is slightly more accurate than the ones...
Currently displaying 61 –
80 of
149