Related Experiment Video
Updated: May 26, 2026
07:05
Applying Hyperspectral Reflectance Imaging to Investigate the Palettes and the Techniques of Painters
Published on: June 18, 2021
Adaptive marginal median filter for colour images
Samuel Morillas1, Valentín Gregori, Almanzor Sapena
1Instituto Universitario de Matemática Puray Aplicada, Universidad Politécnica de Valencia, Camino de Vera s/n, 46022 Valencia, Spain. smorillas@mat.upv.es
Sensors (Basel, Switzerland)
|December 14, 2011
Summary
A novel filter enhances impulse noise reduction in color images by adaptively processing pixels, outperforming the standard vector median filter without introducing color artifacts.
Area of Science:
- Digital Image Processing
- Computer Vision
Background:
- Impulse noise significantly degrades color image quality.
- Classical vector median filters offer limited noise reduction capabilities.
Purpose of the Study:
- To introduce a new adaptive filter for impulse noise reduction in color images.
- To improve upon the noise reduction performance of the vector median filter.
- To ensure the proposed filter avoids introducing color artifacts.
Main Methods:
- A novel filter inspired by vector marginal median filtering.
- Adaptive pixel selection within filtering windows based on the vector median.
- Application of marginal median operation for robust filter design.
Main Results:
- The proposed filter demonstrates superior impulse noise reduction compared to the vector median filter.
- Filtered images exhibit a reduction in noisy pixels.
- The method effectively processes color images without generating color artifacts.
Conclusions:
- The developed adaptive filter offers enhanced impulse noise reduction for color images.
- The filter design is more robust due to its adaptive nature.
- This method provides a viable solution for preserving color fidelity during noise reduction.