Parametric Curve Segment Extraction by Support Regions

TL;DR

This paper presents a novel method for extracting parametric curve segments from images by leveraging the Laplacian of Gaussian filter response and Fourier series modeling of support region boundaries.

Contribution

The method introduces a way to directly obtain convex and concave parametric curve segments from images using support regions and Fourier series modeling.

Findings

Effective extraction of curve segments in parametric form

Handles both convex and concave curves

Uses Fourier series for boundary modeling

Abstract

We introduce a method to extract curve segments in parametric form from the image directly using the Laplacian of Gaussian (LoG) filter response. Our segmentation gives convex and concave curves. To do so, we form curve support regions by grouping pixels of the thresholded filter response. Then, we model each support region boundary by Fourier series and extract the corresponding parametric curve segment.