Catálogo de publicaciones - libros

We describe a novel method of evolutionary visual learning that uses generative approach for assessing learner’s ability to recognize image contents. Each learner, implemented as a genetic programming individual, processes visual primitives that represent local salient features derived from a raw input raster image. In response to that input, the learner produces partial reproduction of the input image, and is evaluated according to the quality of that reproduction. We present the method in detail and verify it experimentally on the real-world task of recognition of hand-drawn shapes.

This paper presents a linear genetic programming approach, that solves simultaneously the region selection and feature extraction tasks, that are applicable to common image recognition problems. The method searches for optimal regions of interest, using texture information as its feature space and classification accuracy as the fitness function. Texture is analyzed based on the gray level cooccurrence matrix and classification is carried out with a SVM committee. Results show effective performance compared with previous results using a standard image database.

We propose a BCI mouse and speller based on the manipulation of P300 waves in EEG signals. The 2–D motion of the pointer on the screen is controlled by directly combining the amplitudes of the output produced by a filter in the presence of different stimuli. This filter and the features to be combined within it are optimised by a GA.

The study of human brain functions has dramatically increased in recent years greatly due to the advent of Functional Magnetic Resonance Imaging. This paper presents a genetic programming approach to the problem of classifying the instantaneous cognitive state of a person based on his/her functional Magnetic Resonance Imaging data. The problem provides a very interesting case study of training classifiers with extremely high dimensional, sparse and noisy data. We apply genetic programming for both feature selection and classifier training. We present a successful case study of induced classifiers which accurately discriminate between cognitive states produced by listening to different auditory stimuli.

This article proposes a Memetic Differential Evolution (MDE) for designing digital filters which aim at detecting defects of the paper produced during an industrial process. The MDE is an adaptive evolutionary algorithm which combines the powerful explorative features of Differential Evolution (DE) with the exploitative features of two local searchers. The local searchers are adaptively activated by means of a novel control parameter which measures fitness diversity within the population. Numerical results show that the DE framework is efficient for the class of problems under study and employment of exploitative local searchers is helpful in supporting the DE explorative mechanism in avoiding stagnation and thus detecting solutions having a high performance.

The is a process by which the 3D point position can be calculated from two images where that point is visible. This process requires the intersection of two known lines in the space. However, in the presence of noise this intersection does not occur, then it is necessary to estimate the best approximation. One option towards achieving this goal is the usage of evolutionary algorithms. In general, evolutionary algorithms are very robust optimization techniques, however in some cases, they could have some troubles finding the global optimum getting trapped in a local optimum. To overcome this situation some authors suggested removing the local optima in the search space by means of a single-objective problem to a multi-objective transformation. This process is called . In this paper we successfully apply this to the triangulation problem.

This paper describes a linear genetic programming approach to multi-class image recognition problems. A new fitness function is introduced to approximate the true feature space. The results show that this approach outperforms the basic tree based genetic programming approach on all the tasks investigated here and that the programs evolved by this approach are easier to interpret. The investigation on the extra registers and program length results in heuristic guidelines for initially setting system parameters.

Feature extraction is a crucial step for Computer Vision applications. Finding appropriate features for an application often means hand-crafting task specific features with many parameters to tune. A generalisation to other applications or scenarios is in many cases not possible. Instead of engineering features, we describe an approach which uses Genetic Programming to generate features automatically. In addition, we do not predefine the dimension of the feature vector but pursue an iterative approach to generate an appropriate number of features. We present this approach on the problem of texture classification based on co-occurrence matrices. Our results are compared to those obtained by using seven Haralick texture features, as well as results reported in the literature on the same database. Our approach yielded a classification performance of up to 87% which is an improvement of 30% over the Haralick features. We achieved an improvement of 12% over previously reported results while reducing the dimension of the feature vector from 78 to four.

We use a genetic algorithm to solve the problem, widely treated in the specialized literature, of fitting an ellipse to a set of given points. Our proposal uses as the objective function the minimization of the sum of orthogonal Euclidean distances from the given points to the curve; this is a non-linear problem which is usually solved using the minimization of the quadratic distances that allows to use the gradient and the numerical methods based on it, such as Gauss-Newton. The novelty of the proposed approach is that as we are using a GA, our algorithm does not need initialization, and uses the Euclidean distance as the objective function. We will also show that in our experiments, we are able to obtain better results than those previously reported. Additionally our solutions have a very low variance, which indicates the robustness of our approach.

The removal of periodic noise is an important problem in image processing. To avoid using the time-consuming methods that require Fourier transform, a simple and efficient spatial filter based on soft mathematical morphology (MM) is proposed in this paper. The soft morphological filter (Soft MF) is optimized by an improved particle swarm optimizer with passive congregation (PSOPC) subject to the least mean square error criterion. The performance of this new filter and its comparison with other commonly used filters are also analyzed, which shows that it is more effective in reducing both periodic and non-periodic noise meanwhile preserving the details of the original image.