EUVIMEDEUVIMED entdecken
Crossref

Adaptive and Optimized RGB Channel Filtering for Noise Reduction in Color Images

Abdelbary M. Algamel

Journal of Technology Research · 25.12.2025 · 3 (Special Issue) · 558-569

DOI: 10.26629/jtr.2025.52 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Technology Research, online erschienen am 25.12.2025. Verfasst von Abdelbary M. Algamel.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper introduces an adaptive filtering method for color image denoising that employs Particle Swarm Optimization (PSO) to tune filter parameters. Building upon a standard mean filter baseline, the approach uses PSO to independently adjust coefficients for each RGB channel according to the specific characteristics of both the image and the noise present. The optimization process targets dual objectives: minimizing Normalized Mean Squared Error (NMSE) and maximizing Signal to Noise Ratio (SNR), while maintaining the core properties of mean filtering. Experimental validation using controlled Gaussian noise demonstrates substantial performance gains: NMSE decreased from 4.81% to 1.11%, and SNR improved from 13.18 dB to 19.55 dB, representing approximately four-fold enhancement over conventional mean filtering. The method shows particular promise for applications demanding image specific optimization, though its effectiveness varies with image content and noise characteristics.

Zitieren

APA

Algamel, A. M. (2025). Adaptive and Optimized RGB Channel Filtering for Noise Reduction in Color Images. Journal of Technology Research, 3(Special Issue), 558-569. https://doi.org/10.26629/jtr.2025.52

Vancouver

Algamel AM. Adaptive and Optimized RGB Channel Filtering for Noise Reduction in Color Images. Journal of Technology Research. 2025;3(Special Issue):558-569. doi:10.26629/jtr.2025.52

BibTeX
@article{Algamel2025,
  author = {Algamel, Abdelbary M.},
  title = {Adaptive and Optimized RGB Channel Filtering for Noise Reduction in Color Images},
  journal = {Journal of Technology Research},
  year = {2025},
  volume = {3},
  number = {Special Issue},
  pages = {558--569},
  doi = {10.26629/jtr.2025.52}
}
Kostenlos starten →