Fuzzy Eigenvector Image Smoothing Method
Khleef Almutairi, Samuel Morillas and Pedro Latorre-Carmona. «Fuzzy Inference Systems to Fine-Tune a Local Eigenvector Image Smoothing Method», published in Electronics 2024, 13, 1150.
This paper introduces a new method for removing noise from color images. Noisy images are a common problem in photography and can come from various sources. This research proposes a technique that combines fuzzy inference systems (FIS) with a method based on analyzing local eigenvectors.
FuzzyEIG1 and FuzzyEIG2
In this paper, two new methods for denoising images, called FuzzyEIG1 and FuzzyEIG2, are introduced. Both of these methods are based on a previously developed strategy known as EIG that uses eigenvectors.
FuzzyEIG1 and FuzzyEIG2 show competitive performance across various metrics and present a more scalable behaviour. Their main advantage is an increased performance in noise reduction concerning EIG while keeping a good level of detail preservation, although a bit lower than that of EIG.
If you want you can download the FuzzyEIG1 and FuzzyEIG2 methods here:
Although our methods have proven to be effective, we can still improve. In the future, we plan to integrate neural networks to optimize the denoising process even further.
