A FUZZY-LOGIC SYSTEM FOR THE DETECTION AND RECOGNITION OF HANDWRITTENSTREET NUMBERS

Fuzzy logic is applied to the problem of locating and reading street n umbers in digital images of handwritten mail. A fuzzy rule-based syste m is defined that uses uncertain information provided by image process ing and neural network-based character recognition modules to generate multiple hypotheses with associated confidence values for the locatio n of the street number in an image of a handwritten address. The resul ts of a blind test of the resultant system are presented to demonstrat e the value of this new approach. The results are compared to those ob tained using a neural network trained with backpropagation. The fuzzy logic system achieved higher performance rates.