How to reduce the size of image?

I am doing my project in OCR.How to reduce the size of character image without any loss in features.I have tried interpolation methods but certain pixels of the images is loss.Could anybody help me to solve this problem.

Answer 1

[Walter Roberson]

It is not possible to resize smaller and preserve all features. It might be possible to resize smaller and preserve the features you are using, maybe; we would have to know what features you are using in order to think about that.

[Minu]

Now i am using image size of 64x64.using features such as zonal density,Zernike moments,Projection Histogram,distance profile .The main problem is feature vector size is too big .I have take the combination of above features and tried.But when i train the neural network ,i have got an error "out of memory".I have tried pca dimensionality reduction but its not work good.i didnt get efficency during training.so i think about reducing the size of an image.

I have one more problem whenever i tried to train the neural network using same features result is varied.how to solve this?

[Walter Roberson]

I seem to recall training neural networks normally involves randomness as well.