Improved Anisotropic Gaussian Filters

TL;DR

This paper introduces a modified anisotropic Gaussian filter algorithm that enhances precision and robustness in fiber orientation estimation, especially in noisy, low-contrast CT images, demonstrated on synthetic and real-world data.

Contribution

A novel modification to 2D anisotropic Gaussian filters that improves accuracy and noise robustness in fiber orientation estimation.

Findings

More accurate orientation estimation on synthetic fiber images.

Increased robustness to noise in low-contrast images.

Effective application to real-world sheet molding compound images.

Abstract

Elongated anisotropic Gaussian filters are used for the orientation estimation of fibers. In cases where computed tomography images are noisy, roughly resolved, and of low contrast, they are the method of choice even if being efficient only in virtual 2D slices. However, minor inaccuracies in the anisotropic Gaussian filters can carry over to the orientation estimation. Therefore, this paper proposes a modified algorithm for 2D anisotropic Gaussian filters and shows that this improves their precision. Applied to synthetic images of fiber bundles, it is more accurate and robust to noise. Finally, the effectiveness of the approach is shown by applying it to real-world images of sheet molding compounds.