Multifocus watermarking approach based on discrete cosine transform.

Area of Science:

• Computer Vision
• Image Processing
• Signal Processing

Background:

• Image fusion combines multiple source images into a single representative image.
• Source images often contain both relevant and irrelevant information.
• Existing fusion methods can introduce artifacts like blurring or blocking.

Purpose of the Study:

• To propose a new image fusion method using Discrete Cosine Transform (DCT).
• To generate a fused image with a more accurate scene depiction than individual source images.
• To ensure the fused image has optimal quality without distortion or data loss.

Main Methods:

• The proposed method utilizes Discrete Cosine Transform (DCT) for image fusion.
• RGB images are decomposed into R, G, and B channels.
• Variance values of 8x8 blocks are computed, and blocks with maximum variance are selected for fusion.
• Inverse DCT is applied to reconstruct the fused image channels.

Main Results:

• The DCT-based fusion method effectively consolidates information from source images.
• The fused image exhibits enhanced detail and superior quality compared to source images.
• The technique mitigates common fusion artifacts such as blurring and blocking.

Conclusions:

• The proposed DCT-based image fusion technique offers an efficient and effective solution.
• This method produces high-quality fused images with improved representational accuracy.
• Experimental results demonstrate superior performance compared to existing image fusion techniques.