EvoWorkshops2002
5th Evolutionary Computing Workshops
3-5 April 2002
Kinsale, Ireland

Menu
Home
EvoCOP2002
EvoIASP2002
Introduction
Programme
Accepted papers
Committee
EvoSTIM2002

Joint event pages
Registration
Programme overview
All accepted papers
Local information
Accommodation
Travel

Main contacts
EvoWorkshops chair
Stefano Cagnoni
Local chair
Conor Ryan

  EvoWorkshops2002

EvoIASP2002
4th European Workshop on Evolutionary Computation in Image Analysis and Signal Processing


Introduction

Evolutionary algorithms have been shown to be tools which can be used effectively in the development of systems (software or hardware) for image analysis and signal processing in complex domains of high industrial and social relevance.

EvoIASP is the first European event specifically dedicated to the applications of evolutionary computation (EC) to image analysis and signal processing (IASP) and gives European and non-European researchers in those fields, as well as people from industry, an opportunity to present their latest research and to discuss current developments and applications, besides fostering closer future interaction between members of the three scientific communities.

The first editions of the Workshop were held in Göteborg, Sweden (1999), Edinburgh, UK (2000), and Lake Como, Italy (2001).

The workshop is sponsored by EvoNet, the Network of Excellence in Evolutionary Computing, and is one of the activities of EvoIASP, the EvoNet working group on Evolutionary Computation for Image Analysis and Signal Processing. It will be part of EvoWorkshops2002 and will be held in conjunction with EuroGP2002, the European Conference on Genetic Programming.

Topics of interest include, but are not limited to:

  • applications of evolutionary computation to real-life IASP problems
  • evolvable vision and signal processing hardware, evolutionary pattern recognition
  • hybrid architectures for machine vision and signal processing including evolutionary components
  • theoretical developments
  • comparisons between different evolutionary techniques and between evolutionary and non-evolutionary techniques in IASP applications
  • time series analysis by means of EC techniques.

The workshop will consist of: an invited talk, oral and poster sessions with periods for discussion possibly along with software demos and industrial stands.

Authors of the best papers submitted to EvoIASP2002 will be invited to submit an extended version of their work for publication in a special issue on Evolutionary Image Analysis and Signal Processing of the "EURASIP Journal of Applied Signal Processing" (http://asp.hindawi.com/).


Programme

Draft: subject to change

See also:
Programme overview | EuroGP programme | EvoCOP programme | EvoSTIM programme

Wednesday 3 April
0830-1130 Registration
0900-1100 Session 1:
Workshop opening
Image and signal processing and analysis

Session chair: Stefano Cagnoni
Using EAs for Error Prediction in Near Infrared Spectroscopy
Fonlupt C., Cahon S., Robilliard D., El-Ghazali T., Duponchel L.
The Boru Data Crawler for Object Detection Tasks in Machine Vision
Howard D., Roberts S. C., Ryan C.
Medical Image Registration Using Parallel Genetic Algorithms
Fan Y., Jiang T., Evans D. J.
Surface Profile Reconstruction From Scattered Intensity Data Using Evolutionary Strategies
Macías D., Olague G., Méndez E. R.
Prediction and Modelling of the Flow of a Typical Urban Basin through Genetic Programming
Dorado J., Rabuńal J. R., Puertas J., Santos A., Rivero D.
1100-1130 Coffee available
1130-1245 EuroGP Session 1:
Conference opening and invited speaker:
Prof Chrystopher L Nehaniv


Session chair: Conor Ryan
1245-1400 Lunch
1430-1530 Session 2:
Image acquisition and synthesis

Session chair: Daniel Howard
Evolutionary Based Autocalibration From the Fundamental Matrix
Whitehead A, Roth G
Efficiently Computable Fitness Functions for Binary Image Evolution
Ványi R
1530-1630 Coffee available
1600-1710 Session 3:
Evolvable hardware

Image Filter Design with Evolvable Hardware
Sékanina L.
A Dynamic Fitness Function Applied to Improve the Generalisation when Evolving a Signal Processing Hardware Architecture
Třrresen J
Evolutionary Techniques for Minimizing Test Signals Application Time
Corno F., Sonza Reorda M., Squillero G.
1710-1800 Session 4:
EC applications to traffic control

Session chair: Giovanni Squillero
The Prediction Of Journey Times On Motorways Using Genetic Programming
Howard D., Roberts S. C.
Detection Of Incidents On Motorways In Low Flow High Speed Conditions By Genetic Programming
Roberts S. C., Howard D.

Workshop close


Accepted papers

Evolutionary Techniques for Minimizing Test Signals Application Time
Corno F., Sonza Reorda M., Squillero G.

Abstract:
Reducing production-test application time is a key problem for modern industries. Several different hardware solutions have been proposed in the literature to ease such process. However, each hardware architecture must be coupled with an effective test signals generation algorithm. This paper propose an evolutionary approach for minimizing the application time of a test set by opportunely extending it and exploiting a new hardware architecture, named interleaved scan. The peculiarities of the problem suggest the use of a slightly modified genetic algorithm with concurrent populations. Experimental results show the effectiveness of the approach against the traditional ones.

Session:
EvoIASP Session 3: Evolvable hardware: April 3, 1600-1710


Prediction and Modelling of the Flow of a Typical Urban Basin through Genetic Programming
Dorado J., Rabuńal J. R., Puertas J., Santos A., Rivero D.

Abstract:
Genetic Programming (GP) is an evolutionary method that creates computer programs that represent approximate or exact solutions to a problem. This paper proposes an application of GP in hydrology, namely for modelling the effect of rain on the run-off flow in a typical urban basin. The ultimate goal of this research is to design a real time alarm system to warn of floods or subsidence in various types of urban basin. Results look promising and appear to offer some improvement over stochastic methods for analysing river basin systems such as unitary radiographs.

Session:
EvoIASP Session 1: Image and signal processing and analysis: April 3, 0900-1100


Medical Image Registration Using Parallel Genetic Algorithms
Fan Y., Jiang T., Evans D. J.

Abstract:
Registration of medical image data of different modalities and multiple times is an important component of medical image analysis. A variety of robust and accurate voxel-based approaches have been proposed, and mathematically almost all of them are associated with optimization problems that are highly non-linear and non-convex. This article presents a parallel genetic strategy to attack mutual information based registration. The experimental results show robust registration with high speedup achieved. Furthermore, this method is readily applicable for other voxel-based registration methods.

Session:
EvoIASP Session 1: Image and signal processing and analysis: April 3, 0900-1100


Using EAs for Error Prediction in Near Infrared Spectroscopy
Fonlupt C., Cahon S., Robilliard D., El-Ghazali T., Duponchel L.

Abstract:
This paper presents an evolutionary approach to estimate the sugar concentration inside compound bodies based on spectroscopy measurements. New European regulation will shortly forbid the use of established chemical methods based on mercury to estimate the sugar concentration in sugar beet. Spectroscopy with a powerful regression technique called PLS (Partial Least Squares) may be used instead. We show that an evolutionary approach for selecting relevant wavelengths before applying PLS can lower the error and decrease the computation time. It is submitted that the results support the argument for replacing the PLS scheme with a GP technique.

Session:
EvoIASP Session 1: Image and signal processing and analysis: April 3, 0900-1100


The Prediction Of Journey Times On Motorways Using Genetic Programming
Howard D., Roberts S. C.

Abstract:
Considered is the problem of reliably predicting motorway journey times for the purpose of providing accurate information to drivers. This proof of concept experiment investigates: (a) the practicalities of using a Genetic Programming (GP) method to model/forecast motorway journey times; and (b) different ways of obtaining a journey time predictor. Predictions are compared with known times and are also judged against a collection of naive prediction formulae. A journey time formula discovered by GP is analysed to determine its structure, demonstrating that GP can indeed discover compact formulae for different traffic situations and associated insights. GP's flexibility allows it to self-determine the required level of modelling complexity.

Session:
EvoIASP Session 4: EC applications to traffic control: April 3, 1710-1800


The Boru Data Crawler for Object Detection Tasks in Machine Vision
Howard D., Roberts S. C., Ryan C.

Abstract:
A 'data crawler' is allowed to meander around an image deciding what it considers to be interesting and laying down flags in areas where its interest has been aroused. These flags can be analysed statistically as if the image was being viewed from afar to achieve object recognition. The guidance program for the crawler, the program which excites it to deposit a flag and how the flags are combined statistically, are driven by an evolutionary process which has as objective the minimisation of misses and false alarms. The crawler is represented by a tree-based Genetic Programming (GP) method with fixed architecture Automatically Defined Functions (ADFs). The crawler was used as a post-processor to the object detection obtained by a Staged GP method, and it managed to appreciably reduce the number of false alarms on a real-world application of vehicle detection in infrared imagery.

Session:
EvoIASP Session 1: Image and signal processing and analysis: April 3, 0900-1100


Surface Profile Reconstruction From Scattered Intensity Data Using Evolutionary Strategies
Macías D., Olague G., Méndez E. R.

Abstract:
We present a study of rough surface inverse scattering problems using evolutionary strategies. The input data consists of far-field angle-resolved scattered intensity data, and the objective is to reconstruct the surface profile function that produced the data. To simplify the problem, the random surface is assumed to be one-dimensional and perfectly conducting. The optimum of the fitness function is searched using the evolutionary strategies (µ ?) and (µ + ?). On the assumption that some knowledge about the statistical properties of the unknown surface profile is given or can be obtained, the search space is restricted to surfaces that belong to that particular class. In our case, as the original surface, the trial surfaces constitute realizations of a stationary zero-mean Gaussian random process with a Gaussian correlation function. We find that, for the conditions and parameters employed, the surface profile can be retrieved with high degree of confidence. Some aspects of the convergence and the lack of uniqueness of the solution are also discussed.

Session:
EvoIASP Session 1: Image and signal processing and analysis: April 3, 0900-1100


Detection Of Incidents On Motorways In Low Flow High Speed Conditions By Genetic Programming
Roberts S. C., Howard D.

Abstract:
Traditional algorithms which set a lower speed limit on a motorway to protect the traffic against collision with a queue are not successful at detecting isolated incidents late at night, in low flow high speed conditions. The Staged Genetic Programming method is used to detect an incident in this traffic regime. The evolutionary engine automatically decides the time duration for the onset of an incident. This method successfully combines traffic readings from the MIDAS system to predict a variety of late night incidents on the M25 motorway.

Session:
EvoIASP Session 4: EC applications to traffic control: April 3, 1710-1800


Image Filter Design with Evolvable Hardware
Sékanina L.

Abstract:
The paper introduces a new approach to automatic design of image filters for a given type of noise. The approach employs evolvable hardware at simplified functional level and produces circuits that outperform conventional designs. If an image is available both with and without noise, the whole process of filter design can be done automatically, without influence of a designer.

Session:
EvoIASP Session 3: Evolvable hardware: April 3, 1600-1710


A Dynamic Fitness Function Applied to Improve the Generalisation when Evolving a Signal Processing Hardware Architecture
Třrresen J

Abstract:
The paper introduces a new approach to automatic design of image filters for a given type of noise. The approach employs evolvable hardware at simplified functional level and produces circuits that outperform conventional designs. If an image is available both with and without noise, the whole process of filter design can be done automatically, without influence of a designer

Session:
EvoIASP Session 3: Evolvable hardware: April 3, 1600-1710


Efficiently Computable Fitness Functions for Binary Image Evolution
Ványi R

Abstract:
There are applications where a binary image is given and a shape is to be reconstructed from it with some kind of evolutionary algorithms. A solution for this problem usually highly depends on the fitness function. On the one hand fitness function influences the convergence speed of the EA. On the other hand, fitness computation is done many times, therefore the fitness computation itself has to be reasonably fast. This paper tries to define what ``reasonably fast'' means, by giving a definition for the efficiency. A definition alone is however not enough, therefore several fitness functions and function classes are defined, and their efficiencies are examined.

Session:
EvoIASP Session 2: Image acquisition and synthesis: April 3, 1430-1530


Evolutionary Based Autocalibration From the Fundamental Matrix
Whitehead A, Roth G

Abstract:
We describe a new method of achieving autocalibration that uses a stochastic optimization approach taken from the field of evolutionary computing and we perform a number of experiments on standardized data sets that show the effectiveness of the approach. The basic assumption of this method is that the internal (intrinsic) camera parameters remain constant throughout the image sequence, i.e. they are taken from the same camera without varying the focal length. We show that for the autocalibration of focal length and aspect ratio, the evolutionary method achieves comparable results without the implementation complexity of other methods. Autocalibrating from the fundamental matrix is simply transformed into a global minimization problem utilizing a cost function based on the properties of the fundamental matrix and the essential matrix.

Session:
EvoIASP Session 2: Image acquisition and synthesis: April 3, 1430-1530


Committee

EvoIASP Chair:

Stefano Cagnoni, University of Parma, Italy

EvoIASP program Committee (to be confirmed):

Giovanni Adorni (Italy)
Wolfgang Banzhaf (Germany)
Alberto Broggi (Italy)
Stefano Cagnoni (Italy)
Ela Claridge (UK)
Terry Fogarty (UK)
Daniel Howard (UK)
Mario Koeppen (Germany)
Evelyne Lutton (France)
Peter Nordin (Sweden)
Riccardo Poli (UK)
Conor Ryan (Ireland)
Jim Smith (UK)
Giovanni Squillero (Italy)
Andy Tyrrell (UK)
Hans-Michael Voigt (Germany)