Hybrid Sharpening Transformation Approach for Multifocus Image Fusion Using Medical and Nonmedical Images.

Area of Science:

• Image Processing
• Computer Vision
• Signal Processing

Background:

• Multifocus image fusion merges images with varying focal planes to create a comprehensive, informative image.
• Preprocessing, particularly image sharpening, is crucial for enhancing details before fusion.
• Existing sharpening methods often lack effectiveness in highlighting key features and edges.

Purpose of the Study:

• To introduce a novel preprocessing transformation approach for multifocus image fusion.
• To propose a new sharpening technique combining the Laplacian filter (LF) and discrete Fourier transform (DFT).
• To evaluate the effectiveness of the proposed LF+DFT sharpening method in improving image fusion quality.

Main Methods:

• Developed a novel sharpening technique, Laplacian filter (LF) + discrete Fourier transform (DFT), for preprocessing multifocus images.
• Applied LF to identify image discontinuities and DFT to analyze frequency changes, enhancing image features.
• Integrated the proposed sharpening technique with advanced fusion methods like stationary wavelet transform (SWT) and discrete wavelet transform (DWT).
• Validated the method on grayscale, color, non-medical, and medical (CT, MRI) datasets.

Main Results:

• The proposed LF+DFT preprocessing significantly improved multifocus image fusion quality.
• Fusion using SWT with LF+DFT demonstrated superior performance across qualitative and quantitative metrics.
• Quantitative evaluation using eight metrics (RMSE, PSNR, etc.) confirmed the method's superiority, with specific metrics reported for the clock dataset.

Conclusions:

• The novel LF+DFT sharpening technique is highly effective for multifocus image fusion.
• The proposed method enhances feature extraction, detail visibility, and edge sharpness.
• This approach offers a significant advancement in image fusion technology for various applications.