This forum is disabled, please visit https://forum.opencv.org
 | 1 | initial version |
Seems to be a classical texture classification problem. Compute your favorite texture-descriptor, like LBP, HOG or HAAR and follow the texton (a.k.a. bag-of-words) approach (see step 2-4 at http://answers.opencv.org/question/8677/image-comparison-with-a-database/#8686). Side-note: Using keypoint-based features typically don't make much sense for uniform textures, thus using texture-features is more favorable.
