Adaptive Structural Interaction Guided Fractional Perona–Malik Diffusion for Medical Image Enhancement

Despite these developments, image enhancement remains a challenging task, because it requires effective noise removal while preserving edges, textures, and fine structural details. Existing diffusion-based techniques may cause excessive smoothing, which can result in the loss of important image information. To address this issue, the present study proposes an image enhancement framework that integrates the Caputo fractional derivative, Hausdorff Fractal Derivative (HFD), and Discrete Wavelet Transform (DWT) within the Perona–Malik anisotropic diffusion model. A fidelity term is incorporated to maintain structural consistency and improve the reconstruction accuracy during the enhancement process. The mathematical properties of the proposed model are established through convergence and stability analyses. The proposed framework is evaluated using two-dimensional and three-dimensional benchmark datasets, as well as medical images corrupted by Gaussian noise. A parameter sensitivity analysis is performed to identify suitable fractional and fractal orders. The results show that parameter values within the range (0, 1) enhance the image while effectively preserving important structural details. The experimental results demonstrate that the proposed method achieves better quantitative and visual performance than the existing image enhancement methods, with higher Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) values and reasonable computational time. Under Gaussian noise, the proposed method demonstrates improved performance compared with deep learning-based methods, including DnCNN, SU-Net, and RED-CNN. To further assess its generalization capability, the proposed framework is evaluated under Gaussian, salt-and-pepper, speckle, Poisson, and mixed noise conditions. The results show that the proposed method provides effective enhancement under different noise conditions, although its performance varies with the type of noise. Multiple realizations of Gaussian noise, together with statistical analysis based on the mean and standard deviation, further confirm the consistency and reliability of the proposed method. In addition, ROI-based comparative analysis demonstrates improved preservation of fine structures and important image details. Overall, the integration of fractional-order diffusion, fractal-based feature preservation, and multiscale wavelet analysis provides an effective and versatile framework for image enhancement.

Authors

Institutions

Publication Details

Journal
Scientific Reports
Published
2026-11-19
DOI
https://doi.org/10.1038/s41598-026-72978-x
Primary Topic
Image and Signal Denoising Methods
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Adaptive Structural Interaction Guided Fractional Perona–Malik Diffusion for Medical Image Enhancement

S. Kalaivani, Priya A
Scientific Reports
Image and Signal Denoising Methods
article

Adaptive Structural Interaction Guided Fractional Perona–Malik Diffusion for Medical Image Enhancement

S. Kalaivani, Priya A
article en

Abstract

Despite these developments, image enhancement remains a challenging task, because it requires effective noise removal while preserving edges, textures, and fine structural details. Existing diffusion-based techniques may cause excessive smoothing, which can result in the loss of important image information. To address this issue, the present study proposes an image enhancement framework that integrates the Caputo fractional derivative, Hausdorff Fractal Derivative (HFD), and Discrete Wavelet Transform (DWT) within the Perona–Malik anisotropic diffusion model. A fidelity term is incorporated to maintain structural consistency and improve the reconstruction accuracy during the enhancement process. The mathematical properties of the proposed model are established through convergence and stability analyses. The proposed framework is evaluated using two-dimensional and three-dimensional benchmark datasets, as well as medical images corrupted by Gaussian noise. A parameter sensitivity analysis is performed to identify suitable fractional and fractal orders. The results show that parameter values within the range (0, 1) enhance the image while effectively preserving important structural details. The experimental results demonstrate that the proposed method achieves better quantitative and visual performance than the existing image enhancement methods, with higher Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index Measure (SSIM) values and reasonable computational time. Under Gaussian noise, the proposed method demonstrates improved performance compared with deep learning-based methods, including DnCNN, SU-Net, and RED-CNN. To further assess its generalization capability, the proposed framework is evaluated under Gaussian, salt-and-pepper, speckle, Poisson, and mixed noise conditions. The results show that the proposed method provides effective enhancement under different noise conditions, although its performance varies with the type of noise. Multiple realizations of Gaussian noise, together with statistical analysis based on the mean and standard deviation, further confirm the consistency and reliability of the proposed method. In addition, ROI-based comparative analysis demonstrates improved preservation of fine structures and important image details. Overall, the integration of fractional-order diffusion, fractal-based feature preservation, and multiscale wavelet analysis provides an effective and versatile framework for image enhancement.

Scientific ReportsVol. 16(1)
Vellore Institute of Technology University (IN)
Image and Signal Denoising Methods
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon