Senior Thesis Research Published: Evolution of Wavelets and Hardware for Satellite Image Compression¶
Allyn Loyd's senior thesis research at Rose-Hulman Institute of Technology has been published as a paper in the proceedings of the International Joint Conference on Computational Intelligence (IJCCI 2025), published by Springer Nature Switzerland.
About the Paper¶
Title: Evolution of Wavelets and Hardware for Satellite Image Compression
Authors: Allyn Loyd, Jason A. Yoder
Venue: International Joint Conference on Computational Intelligence, pp. 442--457, Springer Nature Switzerland, 2025.
The paper explores the use of evolutionary and coevolutionary algorithms to co-optimize wavelets and hardware multipliers for satellite image compression using the discrete wavelet transform (DWT). Designing systems to transmit satellite images requires balancing high compression rates with accurate reconstruction, and the DWT can be fine-tuned for specific image classes but demands substantial hardware resources for floating-point multiplications.
Three algorithmic approaches were investigated:
- A two-stage algorithm that evolves wavelets and hardware separately
- A paired algorithm that evolves them together
- A cooperative coevolutionary algorithm that maintains two separate populations of wavelets and hardware
The results indicate that problem decomposition is most beneficial when evolution allows the components to interact during the search process.
Congratulations to Allyn on this achievement!