Rogério M. Gomes

Automatic fish species classification based on robust feature extraction techniques and artificial immune systems

This paper addresses the problem of automatic classification of fish species, by using image analysis techniques and artificial immune systems. Unlike most common methodologies, which are based on manual estimations that lead to significant time and financial constraints, we present an automatic framework based on (i) two well-known robust feature extraction techniques: Scale-Invariant Feature Transform and Principal Component Analysis for parameterizing shape, appearance and motion, (ii) two immunological algorithms: Artificial Immune Network and Adaptive Radius Immune Algorithm for clustering individuals of the same species, and (iii) a simple nearest neighbor classification strategy. The framework was successfully validated with images of fish species that have significant economic impact, achieving overall accuracy as high as 92%.

A model for hierarchical associative memories via dynamically coupled GBSB neural networks

Many approaches have emerged in the attempt to explain the memory process. One of which is the Theory of Neuronal Group Selection (TNGS), proposed by Edelman [1]. In the present work, inspired by Edelman ideas, we design and implement a new hierarchically coupled dynamical system consisting of GBSB neural networks. Our results show that, for a wide range of the system parameters, even when the networks are weakly coupled, the system evolve towards an emergent global associative memory resulting from the correlation of the lowest level memories.

Information storage and retrieval analysis of hierarchically coupled associative memories

This paper presents information storage and retrieval analysis as well as energy analysis of a multi-level or hierarchically coupled associative memory based on coupled generalised-brain-state-in-a-box (GBSB) neural networks. In this model, the memory processes are described as being organised functionally in hierarchical levels where higher levels coordinate sets of functions of the lower levels. We consider the case where lowest level subnetworks have predefined attractors, prior to imposing their association through imprinting synapses between them. Simulations are carried out using linearly independent (Li) and orthogonal vectors considering a wide range of parameters. The results obtained show that, even when the neural networks are weakly coupled, the system still presents a significant convergence to global patterns, mainly in orthogonal vectors.

Space-vector structure based synthesis for hierarchically coupled associative memories

The present work introduces a proposal for the training of hierarchically coupled associative memories. The method is based on the eigenvalue and eigenvector structure of the space-vector and on suitable changes the space basis. The approach shows to be useful to the class of models hierarchically coupled associative memories, which has the memorization process organized in many levels of degreesof- freedom and for which the training behaves as a synthesis of previously desired states.

Evaluating cluster detection algorithms and feature extraction techniques in automatic classification of fish species

This paper proposes five different schemes for automatic classification of fish species. These schemes make the species recognition based on image sample analysis. Different techniques have been combined for building the classifiers: three feature extraction techniques (PCA, SIFT and SIFT + VLAD + PCA), three data clustering algorithms (aiNet, ARIA and k-means) and three input classifiers (k-NN, SIFT class. and k-means class) are tested. When compared to common methodologies, which are based on human observation, it is believed that these schemes are able to provide significant improvement in time and financial resources spent in classification. Two datasets have been considered: (1) a dataset with image samples of six fish species which are perfectly conserved in formaldehyde solution, and; (2) a dataset composed of images of four fish species in real-world conditions (in vivo). The five proposed schemes have been evaluated in both datasets, and a ranking for the methods has been derived for each one.

An adaptive genetic algorithm to solve the Single Machine Scheduling Problem with Earliness and Tardiness Penalties

This paper deals with the Single Machine Scheduling Problem with Earliness and Tardiness Penalties, considering distinct time windows and sequence-dependent setup time. Due to the complexity of this problem, an adaptive genetic algorithm is proposed for solving it. Many search operators are used to explore the solution space where the choice probability for each operator depends on the success in a previous search. The initial population is generated by applying GRASP to five dispatch rules. For each individual generated, a polynomial time algorithm is used to determine the initial optimal processing date for each job. During the evaluation process, the best individuals produced by each crossover operator, in each generation undergo refinement in order to improve quality of individuals. Computational results show the effectiveness of the proposed algorithm.

Emergence of Neuronal Groups on a Self-Organized Spiking Neurons Network Based on Genetic Algorithm

Based on the Theory of Neuronal Group Selection (TNGS), proposed by Edelman, a network composed of one hundred Izhikevich spiking neurons is analyzed. In this study, a genetic algorithm is used to estimate the Izhikevich neuron model parameters in order to enable the self-organization of a neural network into a cluster of tightly coupled neural cells which fire and oscillate in synchrony at a predefined frequency.

Storage capacity of hierarchically coupled associative memories

This paper, taking as inspiration the ideas proposal for the TNGS (Theory of Neuronal Group Selection), presents a study of convergence capacity of two-level associative memories based on coupled Generalized-Brain-State-in-a-Box (GBSB) neural networks. In this model, the memory processes are described as being organized functionally in hierarchical levels, where the higher levels would coordinate sets of function of the lower levels. Simulations were carried out to illustrate the behaviour of the capacity of the system for a wide range of the system parameters considering linearly independent (LI) and orthogonal vectors. The results obtained show the relations amongst convergence, intensity and density of coupling.

Emergence of synchronicity in a self-organizing spiking neuron network: an approach via genetic algorithms

Based on the Theory of Neuronal Group Selection (TNGS), we have investigated the emergence of synchronicity in a network composed of spiking neurons via genetic algorithm. The TNGS establishes that a neuronal group is the most basic unit in the cortical area of the brain and, as a rule, it is not formed by a single neuron, but by a cluster of tightly coupled neural cells which fire and oscillate in synchrony at a predefined frequency. Thus, this paper describes a method of tuning the parameters of the Izhikevich spiking neuron model through genetic algorithm in order to enable the self-organization of the neural network. Computational experiments were performed considering a network composed of neurons of the same type and another composed of neurons of different types.

Synthesis of frequency generator via spiking neurons network: A genetic algorithm approach

Inspired by the Theory of Neuronal Group Selection (TNGS), we have carried out synthesis of frequency generator via spiking neurons network through genetic algorithm. The TNGS sets that a neuronal group is the most basic unit in the cortical area and are generated by synapses of localized neural cells in the cortical area of the brain firing and oscillating in synchrony at a predefined frequency. Each one of these clusters (Neuronal Groups) is a set of localized, tightly coupled neurons developed in the embryo. According to this proposal, this paper describes a method of tuning the parameters of the Izhikevich spiking neuron model. Computational experiments consisting of a network with all neurons of the same type and a network with different neurons were conducted. A genetic algorithm was used to tune the parameters in these two different cases. The results were compared in order to find the best way to create a frequency generator of spiking neurons network.