Feature classification

I am working on Face recognition and have generated the features using PCA and then LDA. While the features are given as input to the neural net using back propogation, i dont know what to set as target. i tried using the following target but my results are not accurate. e.g for 3 subjects with 3 sample data for each. i set the target as [1 1 1 0 0 0 0 0 0 ; 0 0 0 1 1 1 0 0 0; 0 0 0 0 0 0 1 1 1];

Sumone plz help me out..

[Greg Heath]

The target looks OK to me. Therefore, it is probably something else.

Greg