Michal Bidlo

Evolutionary Design of Arbitrarily Large Sorting Networks Using Development

An evolutionary algorithm is combined with an application-specific developmental scheme in order to evolve efficient arbitrarily large sorting networks. First, a small sorting network (that we call the embryo) has to be prepared to solve the trivial instance of a problem. Then the evolved program (the constructor) is applied on the embryo to create a larger sorting network (solving a larger instance of the problem). Then the same constructor is used to create a new instance of the sorting network from the created larger sorting network and so on. The proposed approach allowed us to rediscover the conventional principle of insertion which is traditionally used for constructing large sorting networks. Furthermore, the principle was improved by means of the evolutionary technique. The evolved sorting networks exhibit a lower implementation cost and delay.

A method for design of impulse bursts noise filters optimized for FPGA implementations

This paper deals with the evolutionary design of area-efficient filters for impulse bursts noise which is often present in remote sensing images such as satellite images. Evolved filters require much smaller area in the FPGA than conventional filters. Simultaneously, they exhibit at least comparable filtering capabilities with respect to conventional filters. Low-cost embedded systems equipped with low-end FPGAs represent a target application for presented filters.

Gate-Level Evolutionary Development Using Cellular Automata

In this paper we present a novel evolutionary developmental technique for the design of the combinational circuits. This technique is based on the development one dimensional uniform cellular automaton. The goal is to evolve a cellular automaton - its local transition function and two different initial states from which a combinational circuit with a given functionality at the gate-level may be developed. The two evolved initial states are intended to demonstrate the ability of the developmental process to construct the given circuit by means of a single local transition function. Moreover, it will be shown that the developmental process is able to adapt also to other initial states than that were originally evolved, i.e. a working circuit possessing a different structure is created. The circuit functionality may be preserved even if the development of the cellular automaton continues after the original circuit was developed.

Evolutionary design of efficient and robust switching image filters

This paper proposes an evolutionary approach based on Cartesian Genetic Programming to the design of image filters for impulse burst noise. The impulse burst noise belongs to more serious image distortions that cause a loss of information in a series of pixels together. The results introduced herein represent a continuation of our research in the design of high-quality image filters. Whilst the previous experiments considered only basic impulse burst noise in which a burst corrupting a series of pixels could take a single value, this paper is devoted to the filtering of more realistic noise of this type where the pixels in a burst can take different values. In order to increase the probability of removing the noise pixels while retaining other pixels unchanged, the concept of switching filter will be applied. In our case it means that the filter system is designed by evolution of both a filter circuit and a noise detector. We show that the proposed method is able to design an efficient and robust impulse burst noise filter that exhibits better filtering properties in comparison with several conventional approaches and, moreover, it is also suitable for a high-speed image processing.

Evolution of cellular automata using instruction-based approach

This paper introduces a method of encoding cellular automata local transition function using an instruction-based approach and their design by means of genetic algorithms. The proposed method represents an indirect mapping between the input combinations of states in the cellular neighborhood and the next states of the cells during the development steps. In this case the local transition function is described by a program (algorithm) whose execution calculates the next cell states. The objective of the program-based representation is to reduce the length of the chromosome in case of the evolutionary design of cellular automata. It will be shown that the instruction-based development allows us to design complex cellular automata with higher success rate than the conventional table-based method especially for complex cellular automata with more than two cell states. The case studies include the replication problem and the problem of development of a given pattern from an initial seed.

Instruction-based development: From evolution to generic structures of digital circuits

Evolutionary techniques provide powerful tools to design novel solutions for hard problems in different areas. However, the problem of scale (i.e. how to create a large, complex solution) represents a significant obstacle for the evolution of complex extensive systems. The computational development represents one of the approaches in the evolutionary design techniques that tries to overcome the problem of scale. In this paper an instruction-based developmental method is presented for the evolutionary design of generic structures of digital circuits. The developmental system involves a set of application-specific instructions constituting programs in order to solve a given task. In particular, the goal is to construct generic structures of combinational circuits. An evolutionary algorithm is utilized for the design of these programs that represent a mapping from the genotypes to the phenotypes during the evolutionary process, i.e. the prescription for the construction of target circuits. Two case-studies are presented in order to demonstrate the successfulness of this approach: (1) the evolutionary design of generic combinational multipliers and (2) the evolutionary design of generic sorting networks.

Evolution of Impulse Bursts Noise Filters

The paper deals with evolutionary design of impulse burst noise filters. As proposed filters utilize the filtering window of 5x5 pixels, the design method has to be able to manage 25 eight-bit inputs. The large number of inputs results in an evolutionary algorithm not able to produce reasonably working filters because of the so-called scalability problem of evolutionary circuit design. However, the filters are designed using an extended version of Cartesian Genetic Programming which enables to reduce the number of inputs by selecting the most important of them. Experimental evaluation of the method has shown that evolved filters exhibit better results than conventional solutions based on various median filters.

Providing information from the environment for growing electronic circuits through polymorphic gates

This paper deals with the evolutionary design of programs (constructors) that are able to create (n+2)-input circuits from n-input circuits. The growing circuits are composed of polymorphic gates considered as building blocks. Therefore, the growing circuit can specialize its functionality according to environment which is sensed through polymorphic gates. The work was performed using a simple circuit simulator. We evolved constructors that are able to create arbitrarily large polymorphic even/odd parity circuits and polymorphic sorting networks.

Evolution of cellular automata with conditionally matching rules

This paper introduces a method of representing transition functions for the purposes of evolutionary design of cellular automata. The proposed approach is based on conditions specified in the transition rules that have to be satisfied in order to determine the next state of a cell according to a specific rule. The goal of this approach is to reduce the number of elements needed to represent a transition function while preserving the possibility to specify traditional transition rules known from the conventional table-based representation. In order to demonstrate abilities of the proposed approach, the replication problem and pattern transformation problem in cellular automata will be investigated. It will be shown that the evolution is able to design transition functions for non-trivial behavior of two-dimensional cellular automata that perfectly fulfil the specified requirements.

Evolutionary design of robust noise-specific image filters

Evolutionary design has shown as a powerful technique in solving various engineering problems. One of the areas in which this approach succeeds is digital image processing. Image filtering represents a wide topic in 2D signal processing. In this case different types of noise are considered in the filtering process to restore the image quality that has been decreased by changing values of some pixels in the image (e.g. due to the transmission through unreliable lines or in the process of acquiring the image). Impulse noise represents a basic type of non-linear noise typically affecting a single pixel in different regions of the image. In order to eliminate this type noise median filters have usually been applied. However, for higher noise intensity or wide range of the noise values this approach leads to corrupting non-noise pixels as well which results in images that are smudged or lose some details after the filtering process. Therefore, advanced filtering techniques have been developed including a concept of noise detection or iterative filtering algorithms. In case of the high noise intensity, a single filtering step is insufficient to eliminate the noise and obtain a reasonable quality of the filtered image. Therefore, iterative filters have been introduced. In this paper we apply an evolutionary algorithm combined with Cartesian Genetic Programing representation to design image filters for the impulse noise that are able to compete with some of the best conventionally used iterative filters. We consider the concept of noise detection to be designed together with the filter itself by means of the evolutionary algorithm. Finally, it will be shown that if the evolved filter is applied iteratively on the filtered image, a high-quality results can be obtained utilizing lower computational effort of the filtering process in comparison with the conventional iterative filters.