how to use neural network to classify different signs made with hands?(hand gesture recognition)

Hi,i am doing project on hand gesture recognition. i have done pre proessing & feature extraction .feature matrix is of 31*31. i need to use neural network for classification.please tell me what shoud be the size of target vector?is it an identity matrix of 31*31?how to use neural network for this method?i want output '1' when first feature vector is given as an input,likevise for others feature vectors.

Answer 1

[Greg Heath]

For N I-dimensional feature vectors from c classes

 [ I N ] = size(input)

 [ c N ] = size(target)

where the columns of target are c-dimensional columns from the unit matrix eye(c).

The correspondence between the targets and class indices are given by

 target = ind2vec(classindices)

 classindices = vec2ind(target)

Hope this helps.

Thank you for formally accepting my answer

Greg

 help patternnet
 doc patternnet