US 5,212,739 AGranted

Noise tolerant optical character recognition system

AssigneeHewlett-Packard Company

InventorDan S. Johnson

Life of this patent

Estimating the term…

Loading patent…

Abstract

Disclosed is a method of optical character recognition that first segments a graphical page image into word images. The method obtains a set of features by extracting smaller outlines of the dark regions in the word images, and then further dissecting each of the smaller outlines into small sections called micro-features. Micro-features are simply sections of character outlines, therefore, they can easily be extracted from the outlines of an entire word without any knowledge about character segmentation boundaries. Micro-features are extracted from an outline by finding the local extremities of the outline and then defining a micro-feature between each pair of sequential extremities. Once extracted, the micro-features are compared to micro-features from an ideal character in order to classify a character, and convert it into a character code.

Claims13

Dependent claims

A system for optical character recognition comprising:

means for converting a page having plurality of text printed thereon into a graphical image containing a plurality of pixel elements representative of said text;
means for separating said graphical image into a plurality of word images;
means for scanning said word images and for producing a plurality of micro-features for each of said word images comprising
means for separating said word images into a plurality of outlines each defined by a boundary between pixels of different intensity within said word images,
means for locating all extremities of each of said outlines, comprising
means for combining each pair of adjacent edge points of said outlines into a plurality of segments,
means for computing a directional orientation for each of said segments,
means for combining adjacent segments having the same directional orientation into a plurality of sections, and
means for defining zero or more points in each of said sections as an extremity, and
means for producing a single micro-feature between each pair of said extremities, said micro-feature being terminated at each of said pair of extremities;
and
means for converting said micro-features into a set of coded characters equivalent to said word.

9 more elements

A system for optical character recognition comprising:

means for converting a page having plurality of text printed thereon into a graphical image containing a plurality of pixel elements representative of said text;
means for separating said graphical image into a plurality of group images, each said group image having one or more character images;
means for scanning said group images and for producing a plurality of micro-features for each of said group images comprising
means for operating said group image into a plurality of outlines each defined by a boundary between pixels of different intensity within said group image,
means for locating all extremities of each of said outlines, comprising
means for combining each pair of adjacent edge points of said outlines into a plurality of segments,
means for computing a directional orientation for each of said segments,
means for combining adjacent segments having the same directional orientation into a plurality of sections, and
means for defining zero or more points in each of said sections as an extremity, and
means for producing a single micro-feature between each pair of said extremities, said micro-feature being terminated at each of said pair of extremities;
and
means for converting said micro-features into a set of coded characters each equivalent to one of said character images in said group image.

9 more elements

Dependencies are inferred from the claim text.