Robust Image Corner Detection Through Curvature Scale Space

Abstract

Abstract-This paper describes a novel method for image corner detection based on the curvature scale-space (CSS) representation. The first step is to extract edges from the original image using a Canny detector. The corner points of an image are defined as points where image edges have their maxima of absolute curvature. The corner points are detected at a high scale of the CSS and tracked through multiple lower scales to improve localization. This method is very robust to noise, and we believe that it performs better than the existing corner detectors. An improvement to Canny edge detector's response to 45 o and 135 o edges is also proposed. Furthermore, the CSS detector can provide additional point features (curvature zerocrossings of image edge contours) in addition to the traditional corners.

A Motion-Based Feature for Event-Based Pattern Recognition

“…This measure can be computed by a cumulative process (Park et al, 2004), using a self-similarity measure (Moravec, 1980) derived from mathematical analysis [e.g., contour's local curvature (Mokhtarian and Suomela, 1998), relying on an eigenvalue decomposition of a second-moment matrix (Harris and Stephens, 1988)] or selected as the output from a machine learning process (Rosten and Drummond, 2006). …”

Laplacian Scale-Space Behavior of Planar Curve Corners

“…In fact, except for the convolution of a LoG operator and boundary coordinates, which have the same computational burden as the convolution of Gaussian and boundary coordinates, the response function has only two multiplication operations. In other words, these two attributes lead to the lowest computational complexity among the existing classical boundary-based corner detection algorithm, such as CSS [20], wavelet [30], covariance matrices [31] and CPDA [23], etc. In Figure. 13, we present some corner detection results of the proposed corner detection algorithm on several images which are usually used for corner detection.…”

“…Particularly, the scale-space technique has been extensively utilized in corner detection in the past two decades, e.g., [19] [20] [21] [22] [23] [24] [25] [26]. It is necessary to figure out the scale-space behavior of corners.…”

Contour-Based Corner Detection and Classification by Using Mean Projection Transform

“…Γ is considered the curvature at a point, as presented in Equation (9) [44]: $\Gamma =\frac{\text{dψ}}{\text{ds}}$where ψ is the change rate of angle, and the corresponding S can be defined as the arc-length. Curve smoothing reduces sensitivity to the local variation of the contour [20]. The CSS-based detectors, which use contour smoothing, are not sufficient to detect the corner points based on their evaluation results.…”