A Novel Approach for Interest Point Detection via Laplacian-of-Bilateral Filter

Algorithm 1

Proposed interest point detection algorithm.

(1) To detect extrema at scales of each octave, compute Bilateral response for all scales ,

  using (7) with .

(2) Compute the Laplacian-of-Bilateral response by using (11).

(3) Extract the maximal/minimal point in scale-space of the response in a window as shown in Figure 3.

(4) Remove edge responses and low contrast points using Hessian matrix similar to SIFT algorithm.

(5) Assign orientation and compute the histogram of magnitude of first order gradients over

  different orientations resulting in a feature descriptor vector of dimension 128 analogous to SIFT algorithm.

(6) Store the feature location, scale, orientation, and feature descriptor.