Catálogo de publicaciones - libros

In this paper we address the problem of aligning two partially overlapping surfaces represented by points obtained in subsequent 2D scans for mobile robot pose estimation. The measured points representation contains incomplete measurements. We solve this problem by minimizing an alignment error via a genetic algorithm. Moreover, we propose an alignment metric based on a look-up table built during the first scan. Experimental results related to the convergence of the proposed algorithm are reported. We compare our approach with other scan matching algorithms proposed in the literature, and we show that our approach is faster and more accurate than the others.

Overcompression is the process of post-processing compressed images to gain either further size reduction or improved quality. This is made possible by the fact that the set of all “reasonable” images has a sparse structure. In this work, we apply this idea to the overcompression of JPEG images: We reduce the blocking artifacts commonly seen in JPEG images by allowing the low frequency coefficients of the DCT to vary slightly. Evolutionary strategies are used in order to guide the modification of the coefficients towards a smoother image.

This paper discusses a study towards dynamic fitness based partitioning in IntraVascular UltraSound (IVUS) image analysis. Mixed-Integer Evolution Strategies (MI-ES) have recently been successfully used to optimize control parameters of a multi-agent image interpretation system for IVUS images lumen detection. However, because of complex interpretation contexts, it is impossible to find one single solution which works well on each possible image of each possible patient. Therefore it would be wise to let MI-ES find a of solutions based on an optimal partition of IVUS images. Here a methodology is presented which does dynamic fitness based partitioning of the data during the MI-ES parameter optimization procedure. As a first step we applied this method to a challenging artificial test case which demonstrates the feasibility of our approach.

A Hidden Markov Model (HMM) is used as an efficient and robust technique for human activities classification. The HMM evaluates a set of video recordings to classify each scene as a function of the future, actual and previous scenes. The probabilities of transition between states of the HMM and the observation model should be adjusted in order to obtain a correct classification. In this work, these matrixes are estimated using the well known Baum-Welch algorithm that is based on the definition of the real observations as a mixture of two Gaussians for each state. The application of the GA follows the same principle but the optimization is carried out considering the classification. In this case, GA optimizes the Gaussian parameters considering as a fitness function the results of the classification application. Results show the improvement of GA techniques for human activities recognition.

This work describes an evolutionary approach to texture segmentation, a long-standing and important problem in computer vision. The difficulty of the problem can be related to the fact that real world textures are complex to model and analyze. In this way, segmenting texture images is hard to achieve due to irregular regions found in textures. We present our algorithm, which uses knowledge derived from texture analysis to identify how many homogeneous regions exist in the scene without information. uses texture features derived from the Gray Level Cooccurrence Matrix and optimizes a fitness measure, based on the minimum variance criteria, using a hierarchical GA. We present qualitative results by applying on synthetic and real world images and compare it with the state-of-the-art JSEG algorithm.

Photographic supra-projection is a complex and uncertain process that aims at identifying a person by overlaying a photograph and a model of the skull found. The more accurate the skull model is the more reliable the identification decision will be. Usually, forensics are obliged to perform a manual and time consuming process in order to obtain the model of the scanned forensic object. At least semiautomatic methods are demanded by these experts to assist them with this task. Our contribution aims to propose an evolutionary-based image registration methodology for the skull 3D model building problem. Experiments are performed over thirty two problem instances corresponding to a semiautomatic and fully automatic real skull model reconstruction.

This work presents scale invariant region detectors that apply evolved operators to extract an interest measure. We evaluate operators using their repeatability rate, and have experimentally identified a plateau of local optima within a space of possible interest operators Ω. The space Ω contains operators constructed with Gaussian derivatives and standard arithmetic operations. From this set of local extrema, we have chosen two operators, obtained by searching within Ω using Genetic Programming, that are optimized for high repeatability and global separability when imaging conditions are modified by a known transformation. Then, by embedding the operators into the linear scale space generated with a Gaussian kernel we can characterize scale invariant features by detecting extrema within the scale space response of each operator. Our scale invariant region detectors exhibit a high performance when compared with state-of-the-art techniques on standard tests.

Design and optimisation of modern complex mixed analogue-digital circuits require new approaches to circuit sizing. In this paper, we present a novel approach based on the application of the univariate marginal distribution algorithm to circuit sizing at the system level. The results of automotive electronics circuits sizing indicate that all design requirements have been fulfilled in comparison with a human design. Experiments indicate that elitism increases the performance of the algorithm.

In this paper, we propose a method to support texture mapping for intuitive designing of 3D objects in the scene of virtual or real space. To fit a texture pattern to the target 3D object, users should consider not only physical constraints, such as resolutions and scales, but also psychological constraints, such as fitness or moods. This method generates multiple candidate models applying various kinds of texture patterns and allows users to evaluate them sensitively. The technique called Interactive Evolutionary Computation (IEC) helps them to find the pleasant and adequate a texture pattern for the scene easily. Texture pattern is improved corresponding to users’ evaluations. This framework provides a powerful environment for interactive texture design.

The research presented in this paper deals concerns interactive evolutionary design systems and specifically with the Interactive Evolutionary Design Environment (IEDE) developed by the authors. We describe the IEDE concentrating upon the three major components: Component-based Representation; Construction and Repair Agents (providing build and evolve services) and a machine learning sub-system. We also describe the clustering technique utilized within the IEDE to improve the user interactivity of the system.