Ask Your Question

Train from color, not from object.

asked 2013-09-27 06:22:11 -0500

updated 2013-09-27 12:12:26 -0500

Is it possible to train an LM with an images folder, using a defined start color for each image ?

I mean :

  1. Start with know color hsv/rgb
  2. On each picture saved, learn modification
  3. Next start(1 month, 1 day, 1 year) use this learned range to predict the new color range
  4. and so on
edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2013-09-27 07:20:36 -0500

updated 2013-09-27 12:53:41 -0500

Yes.For example in the FACE ICAO field for compute of light balance in the face in the first you have to specific the skin region of face.If you do this by controlling the range of colors( Hue,Saturation,Intensity or RGB)by rules the final result is not accurate because our method is linear but our problem is nonlinear.

To obtain an accurate result you can train a classifier with color features for example H-S-I-R-G-B that be specified by the user(for select colors use the flood fill).In the final use the MLP or SVM for classify it.

edit flag offensive delete link more


Ok Mostafa, so I can train the classifier with the histogram of the floodfilled images, and calcBackProject the predicted histogram ? BTW, after some test, I dont know how to setup the response Mat for svm training

Pascal66 gravatar imagePascal66 ( 2013-09-27 16:01:17 -0500 )edit

Question Tools

1 follower


Asked: 2013-09-27 06:22:11 -0500

Seen: 231 times

Last updated: Sep 27 '13