Feature matching of segmented images

asked 2013-03-30 08:02:42 -0600

matteo
153 ●4 ●5 ●9

updated 2013-03-30 08:06:01 -0600

image description hi everybody, i'm using this example proposed by OpencCV for feature extraction from images of segmented roads (one translated with respect to the other). My problem is that it finds 0 matching point: how is it possible? thanks

edit retag flag offensive close merge delete

add a comment

1 answer

Sort by » oldest newest most voted

answered 2013-03-30 08:15:12 -0600

Guanta
6736 ●6 ●25 ●79

updated 2013-03-30 08:16:52 -0600

Binary images don't have enough structure (entropy) for the typical (afaik none) keypoint-based features. Either don't convert to binary and just take the grayscale converted images to feed into your feature-extractor or you have to use a different approach like shape-matching.