April 2020
Intermediate to advanced
438 pages
12h 2m
English
If you run the preceding code snippets and plot the matches starting from the keypoints and descriptors obtained with/without RANSAC algorithm and then plot the second image warped onto the first one using the estimated homography matrix, you will get an output like the following screenshot:

As you can see from the next screenshot, using the RANSAC algorithm removes all of the wrong matches and generates a much better projective transform.
The corner_harris()and corner_peaks() functions from skimage.feature are used to obtain the Harris Corner keypoints from the input images (it finds points with a large corner response function ...
Read now
Unlock full access