Eastman-kodak Company

Andrew M Assad

Roger S Gaborski

Lori L Barski

Yongchun Lee

Louis J Beato

Thaddeus Francis Pawlicki

Hin-Leong Tan

Dawn L Dutton

A character bit map is skeletonized using competitive locally independent processes (CLIPs) by defining for each foreground pixel an individual CLIP with a center location and radius defining the pixels included in the CLIP. In iterative cycles, each CLIP varies its center location and radius over predetermined ranges to produce a predetermined number of permutations thereof, computes a trial fitness value for each permutation, the trial fitness value being rewarded and penalized in accordance with foreground and background pixels included in the CLIP and penalized for each pixel included in both the CLIP and in another CLIP of greater fitness value. Each CLIP adopts the maximum trial fitness value as its next fitness value along with the corresponding center location and radius. Each CLIP is eliminated if its next fitness value falls below a predetermined minimum value. Whenever an iterative cycle fails to produce a change in the number or radii of the CLIPs greater than a predetermined threshold, the center locations of all surviving CLIPs are output as a skeleton character map.

Character thinning using emergent behavior of populations of competitive

Character images which are to be sent to a neural network trained to recognize a predetermined set of symbols are first processed by an optical character recognition pre-processor which normalizes the character images. The output of the neural network is processed by an optical character recognition post-processor. The post-processor corrects erroneous symbol identifications made by the neural network. The post-processor identifies special symbols and symbol cases not identifiable by the neural network following character normalization. For characters identified by the neural network with low scores, the post-processor attempts to find and separate adjacent characters which are kerned and characters which are touching. The touching characters are separated in one of nine successively initiated processes depending upon the geometric parameters of the image. When all else fails, the post-processor selects either the second or third highest scoring symbol identified by the neural network based upon the likelihood of the second or third highest scoring symbol being confused with the highest scoring symbol.

Optical character recognition neural network system for machine-printed

Characters which touch one another in a document image are identified and separated in accordance with their contours. The objects in the image are divided into two classes comprising, respectively, those having closed inner contours and those which do not. An object in the first class having either a closed inner contour which is not centered in the object or two horizontally offset closed inner contours or more than two closed inner contours is identified as containing two touching characters and is split in accordance with the character stroke thickness observed in the document. An object in the second class whose contours intersect three spaced horizontal reference lines at more than a predetermined number of points is compared against intersection patterns characteristic of single characters. If a match is not found, the object is identified as containing touching characters.

Contour feature-based method for identification and segmentation of

Growing community of inventors

N. Chili, NY, United States of America

Andrew M Assad