7.4 Label Recognition Training
We now revisit the Label Recognition example first introduced in the Unsupervised Filter. We must prepare the match.txt file for training. This file is already prepared for you and we will simply open it and save it to match.txt. The steps are:
Match.txt
-
Open the file, “.\data\match_ex_label.txt” (the full path is “c:\program files\Attrasoft\ImageFinder 7.0\data\ match_ex_label.txt”) via Windows Explorer (Right click Start\Explorer). This file lists 152 matching pairs. Save it to match.txt (overwrite the existing file). Now the training file is prepared.
T1.txt:
-
Click the “Source” button, go to “.\ex_label” directory and select any file in the folder. This will specify the input directory.
-
Click the Source “>” button a few times to see the images;
-
Click menu item “Signature/N Signature (a1.txt)” to get signature file, a1.txt file;
-
Click menu item “Signature/Copy a1.txt to t1.txt” to get the training file, t1.txt.
Note: Here t1.txt is for training and a1.txt is for 1:N Matching and N:N Matching.
Training
-
Click “BioFilter\Training\Training” to train the BioFilter.
You should get this message at the end of the text window:
Total Number of Matches = 152
Number of Images that have No Match = 152
There are 304 images in 152 pairs. The match.txt listed 152 pairs. The first line, Total Number of Matches = 152, indicates the training used 152 pairs. The second line, Number of Images that have No Match = 152, indicates 152 out of 304 images does not have a match, which is correct. This is because in match.txt which has 152 pairs, (A, B), only A will match with B, but B will not match with A.
Now, the BioFilter is trained for the Label Recognition problem. We will continue this example in the next section, N:N Matching.
Copyright (c) 1998 - 2006 Attrasoft, Inc. All rights reserved.