Method for recognizing characters

A device for recognizing characters comprising:

• A. means for recording successively the image of each character to be recognized,
• B. means for manipulating each recorded image for scanning it from different directions to detect a predetermined number of aspects of the pattern of each character,
• C. means for weighing each feature detected in each of said aspects of each character,
• D. means for selecting the features having the maximum weights in each processed aspect of each character,
• E. a learning matrix means for storing the resulting selected maximum weights according to the position of the features in the aspect of each character for all the characters to be recognized,
• F. means for adding the values of each weight to be stored to the value of weights already stored for that feature at that location in said learning matrix means, and
• G. means for selecting the maximum of the added results for indicating the class of the character to be recognized.

A device for recognizing characters comprising:

• an image manipulator (9);
• a learning matrix (13);
• a number of processing units (10) for computing the number of jumps, slopes, end points, and islands are features of a character when scanned from four different inside and outside directions as aspects, and determining the weights of these features per aspect depending upon the shape of the pattern of the character to be recognized;
• a number of maximum selectors (11) for choosing, from each group of features per aspect, the feature having the highest weight;
• a number of coding devices (12) for completing, during the learning phase, a store of the learning matrix for each of the highest values;
• a number of normalizing devices (14) for establishing a relation between the value determined by the selectors (11) and the size of the image in the same aspect;
• a number of logic circuits (15) for determining the logarithmic values of the results obtained by the normalizing devices (14);
• a number of summing circuits (16) for adding the logarithmic values obtained from the logic circuits (15) to the logarithmic values of the values recorded per aspect in the learning matrix for the relevant feature in all the character classes;
• a second matrix (17) for storing, for all the character classes, the logarithmic values found in the summing devices (16) and a maximum selector (18) for determining the highest value of the sum of the feature values for all the aspects of all classes of characters.