Autocorrelation-Driven Diffusion Filtering

SND-ID: 2024-219. Version: 1. DOI: https://doi.org/10.5878/qb4q-jt57

Citation

Creator/Principal investigator(s)

Michael Felsberg - Linköping University, Department of Electrical Engineering, Computer Vision / Center for Medical Image Science and Visualization (CMIV) orcid

Research principal

Linköping University rorId

Description

The dataset consists of Matlab code and present a novel scheme for anisotropic diffusion driven by the image autocorrelation function. We show the equivalence of this scheme to a special case of iterated adaptive filtering. By determining the diffusion tensor field from an autocorrelation estimate, we obtain an evolution equation that is computed from a scalar product of diffusion tensor and the image Hessian. We propose further a set of filters to approximate the Hessian on a minimized spatial support. On standard benchmarks, the resulting method performs favorable in many cases, in particular at low noise levels. In a GPU implementation, video real-time performance is easily achieved.

The dataset was originally published in DiVA and moved to SND in 2024.

Data contains personal data

No

Language

Method and outcome
Data collection
Geographic coverage
Administrative information

Identifiers

Topic and keywords

Research area

Computer vision and robotics (autonomous systems) (Standard för svensk indelning av forskningsämnen 2011)

Publications
Published: 2018-01-19
Last updated: 2024-07-01