Related Experiment Video
Updated: Jun 27, 2026
15:25
Design and Characterization Methodology for Efficient Wide Range Tunable MEMS Filters
Published on: February 4, 2018
An enhanced memetic differential evolution in filter design for defect detection in paper production
Ville Tirronen1, Ferrante Neri, Tommi Kärkkäinen
1Department of Mathematical Information Technology, Agora, University of Jyväskylä, P.O. Box 35 (Agora), FI-40014 University of Jyväskylä, Finland. aleator@jyu.fi
Evolutionary Computation
|December 5, 2008
Summary
This study introduces an Enhanced Memetic Differential Evolution (EMDE) algorithm for designing digital filters. The EMDE effectively detects paper defects in industrial processes using Gabor filters, outperforming other metaheuristics.
Area of Science:
- Engineering
- Computer Science
- Image Processing
Background:
- Industrial paper production requires robust defect detection systems.
- Digital filter design is crucial for accurate image analysis and quality control.
- Existing methods may lack efficiency or adaptability in complex industrial settings.
Purpose of the Study:
- To propose and evaluate an Enhanced Memetic Differential Evolution (EMDE) algorithm.
- To design digital filters for detecting defects in industrial paper production.
- To leverage Gabor filters for enhanced defect identification.
Main Methods:
- Developed an adaptive evolutionary algorithm (EMDE) combining Differential Evolution with local search.
- Integrated Hooke Jeeves Algorithm, Stochastic Local Search, and Simulated Annealing.
- Employed a novel control parameter and probabilistic scheme for adaptive coordination.
Main Results:
- EMDE demonstrated strong performance in designing digital filters for defect detection.
- The algorithm showed efficient convergence and effective stagnation prevention.
- Successfully designed tailored filters for industrial paper defect identification.
Conclusions:
- EMDE is an effective evolutionary framework for image processing tasks like defect detection.
- The proposed algorithm offers superior performance compared to various metaheuristics.
- EMDE enables the design of efficiently tailored filters for industrial applications.
