1
$\begingroup$

I am trying to achieve a pattern matching between an image and a scene in which the image is. I am having problems with some images that does not contain corners, since SURF is not able to find keypoints.

Does anybody know any algorithm or method for retrieving keypoints in images without corners? In other words: how can I find an image inside another when this image does not contain corners?

I am using Matlab. Thank you in advance,

$\endgroup$
  • 1
    $\begingroup$ Could you post a sample image? $\endgroup$ – Dima Jan 4 '14 at 21:46
4
$\begingroup$

The computer vision system toolbox in Matlab includes Feature Detection, Extraction, and Matching. The line, edges, histogram of oriented gradients are all included other than the corner features. And there is also an example on point feature matching in the toolbox (the names are called corners but they are actually not).

$\endgroup$
1
$\begingroup$

If you are after the simplest algorithm, you can basically try Genaralized Hough Transform using the image edges only (not corners).

For really textureless objects, the following might guide you:

1) If you are willing to use an RGBD camera such as kinect, Hinterstoisser has come up with a quite good algorithm to do this:

http://cvlabwww.epfl.ch/~lepetit/papers/hinterstoisser_pami11.pdf

2) If not, you can use BOLD features (Bunch of Line detectors). This would assume your template is composed of line features, and not corners. This paper is also quite comfortable to read:

http://vision.deis.unibo.it/fede/papers/iccv13_pre.pdf

Note that the papers I provided are published after LineMod (or Line2D), which is already in OpenCV: http://ar.in.tum.de/pub/hinterstoisser2011linemod/hinterstoisser2011linemod.pdf

$\endgroup$
1
$\begingroup$

It depends which types of the features are useful for you. If the corners are not useful/caracteristic, you should look elsewhere. For example, may be blob-like features will suit you need. Take a look at wikipedia article on feature detectors http://en.wikipedia.org/wiki/Feature_detection_(computer_vision). It is a good starting point and will give you an idea what are the groups and assumptions. Nowadays there are many toolboxes that integrate many of those algorithms. For example, recent MATLAB versions already include toolboxes for computer vision.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.