How to match one image with best one from different images

Hi everybody. I group of images of an object from different positions, and i take new image of this object and want to match it with the best possible image from a group (the same side and position). Thanks.

Answer 1

[Florian Morsch]

Detect SURF features on all your previous compare-images, then detect the features on the newly taken image.

Now check the SURF feature points with each image and take the one which fits best.

Example can be found here: https://de.mathworks.com/help/vision/examples/object-detection-in-a-cluttered-scene-using-point-feature-matching.html

[Baisseyev Miram]

Thanks. I need to select best match manually? is there a way to make it automatically?

[Florian Morsch]

You can search for the two matched with the most shared SURF features.

One possible option: write all shared featurepoints in an array (like when you detect 10 in the first match, 13 in the second, 9 in the third and 22 in the last) your array would look like [10 13 9 22] and then just search for the highest value in the array and take the corresponding picture. You can do so with

[max, index] = max(yourArray(:))

this gives you the largest output from your array and the position in which is stored ( here is the complete link https://de.mathworks.com/help/matlab/ref/max.html ). In this case you would get max = 22, index = 4, so you would know your 4th picture is the best match.