Efficient Image De-Noising Technique Based on Modified Cuckoo Search Algorithm

Area of Science:

• Image Processing
• Computer Vision

Background:

• Image restoration is crucial for enhancing visual quality and recovering original image details.
• Previous noise elimination methods like OGHP and SURE shrinkage showed limitations in PSNR and SSIM efficiencies.

Purpose of the Study:

• To develop an innovative image restoration technique for improved noise reduction and visual augmentation.
• To address limitations of existing methods by incorporating adaptive preprocessing and optimized noise elimination.

Main Methods:

• Introduced salt and pepper and impulse noise to images.
• Applied adaptive median and adaptive fuzzy switching filters for preprocessing.
• Utilized optimized OGHP (Original GHP) with Cuckoo Search Algorithm and SURE shrinkage for noise elimination.
• Employed Discrete Wavelet Transform (DWT) for noise-eliminated image estimation.
• Implemented AGA approach for the final image restoration procedure.

Main Results:

• The novel method demonstrated superior efficiency in image restoration compared to modern techniques.
• Performance outcomes indicated significant success in noise elimination and visual quality enhancement.
• The approach effectively restored images degraded by salt and pepper and impulse noise.

Conclusions:

• The proposed image restoration method offers a significant advancement in the field.
• The combination of adaptive filtering, optimized noise elimination, and AGA approach yields high-quality restored images.
• This technique provides a robust solution for various image noise challenges.