Steps: 1: Computation of the digital Gaussian scale-space

2: Bilinear interpolation of an image

3: Computation of the difference of Gaussians scale-space (DoG)

4: Scanning for 3d discrete extrema of the DoG scale-space

5: Discarding low contrasted candidate keypoints (conservative test)

6: Keypoints interpolation

7: Quadratic interpolation on a discrete DoG sample

8: Discarding low contrasted candidate keypoints

9: Discarding candidate keypoints on edges

10: Computation of the 2d gradient at each image of the scale-space

11: Computing the keypoint reference orientation

12: Construction of the keypoint descriptor