Constrained Regularization Approach to Curvature Estimation

In this paper, we propose a method for boundary smoothing using constrained regularization technique. The smoothed boundary is used to compute a curvature function. Computing a curvature function on a digitized boundary is an ill-posed problem due to the discrete nature of the boundary. Thus, the boundary needs to be smoothed before computing the curvature function. We use a constrained regularization technique to obtain the optimal smooth boundary. The curvature function is computed porn this optimal smooth boundary. This method solves a common critical problem of current curvature estimation methods in determining a unique smoothing factor. Using the above method, we derive a curvature function which is invariant under rotation, scale and translation. Thus, the resulting curvature function can be used to uniquely detect corner points.