15.4 Training Design
About1 Introduction2 Image Recognition3 TransApplet4 API5 Interface6 Input7 Image Display8 Preprocessing9 Processing10 Normalization11 Parameter Class12 Image Signatures13 Unsupervised Filters14 BioFilters15 NeuralFilters16 Dynamic Library17 NeuralNet Filter18 Parameters19 Input Options20 Database Input21 Video Input22  Live Video Input23  Counting & Tracking24  Counting 25  Batch Job26 ImageFinder for DOS27 ImageHunt 28 Support Packages

15.1 NeuralFilter Menu 
15.2 NeuralFilter API 
15.3 Parameters 
15.4 Training Design 
15.5 Implementation 
15.6 N:N Design 
15.7 N:N Implementation 
15.8 1:N Design 
15.9 1:N Implementation 
[Home][15 NeuralFilters][15.4 Training Design]


15.4   Training Design

We introduced the Label Recognition problem with both the Unsupervised Filter and the BioFilter. We will choose “Signature Filter 9” in this chapter. In Figure 15.2, select Signature Filter 9 from line 6.

Figure 15.2  Parameter Window.

We now revisit the Label Recognition example.  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:


  •    Open the file, “.\data\match_ex_label.txt”. This file lists 152 matching pairs. Save it to match.txt (overwrite the existing file). Now the training file is prepared.


  •    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.


  •    Click “Neural Filter\Training\Training” to train the NeuralFilter.

You should get this message at the end of the text window:

      Number of Matches = 152

      Neural Filter Training Completed!

The match.txt listed 152 pairs. The first line, Total Number of Matches = 152, indicates the training used 152 pairs. Now, the Neural Filter is trained for the Label Recognition problem. We will continue this example in the next section, N:N Matching.


[Home][About][1 Introduction][2 Image Recognition][3 TransApplet][4 API][5 Interface][6 Input][7 Image Display][8 Preprocessing][9 Processing][10 Normalization][11 Parameter Class][12 Image Signatures][13 Unsupervised Filters][14 BioFilters][15 NeuralFilters][16 Dynamic Library][17 NeuralNet Filter][18 Parameters][19 Input Options][20 Database Input][21 Video Input][22 Live Video Input][23 Counting & Tracking][24 Counting ][25 Batch Job][26 ImageFinder for DOS][27 ImageHunt ][28 Support Packages]

Copyright (c) 2006 - 2007 Attrasoft. All rights reserved.