I have been making a OCR application for degraded text documents ...i would like to know what is the best algorithm to use to extract the features and word recognition..
I used the HMM method but doesnt seems to give me good results.. info will be much appreciated