Use of convexity in contour detection

TL;DR

This paper introduces a convexity-based algorithm for contour detection in images, utilizing topographic surface modeling and calculus testing, capable of multi-scale analysis and compared favorably to watershed methods.

Contribution

The paper presents a novel convexity-based contour detection algorithm that is simple, adaptable to multiple scales, and offers an alternative to watershed transform methods.

Findings

Effective contour detection demonstrated on multiple images

Multi-scale detection achieved through initial smoothing adjustments

Comparable or improved results relative to watershed transform

Abstract

In this paper, we formulate a simple algorithm that detects contours around a region of interest in an image. After an initial smoothing, the method is based on viewing an image as a topographic surface and finding convex and/or concave regions using simple calculus-based testing. The algorithm can achieve multi-scale contour detection by altering the initial smoothing. We show that the method has promise by comparing results on several images with the watershed transform performed on the gradient images.