System for extracting information from an identity card
First Claim
Patent Images
1. A system for extracting information from an identity document comprising at least one light source for illuminating the identity document;
- at least one digital camera comprising a lens and a two-dimensional sensor array for acquiring at least one two-dimensional image of the identity document; and
a processor for processing said at least one two-dimensional image, said processor configured to run an optical character recognition (OCR) routine for extracting textual information from the identity document in the form of alphanumerical information, wherein said extracting textual information comprises the following steps;
a) determining boundaries between text and background based on minimum contrast between the character and other colored data on the identity document;
b) extracting black and white glyphs using at least a portion of said determined boundaries;
c) applying said OCR routine on said black and white glyphs, thereby recognizing symbols and characters; and
d) sending said recognized symbols and characters as alphanumerical information.
6 Assignments
0 Petitions
Accused Products
Abstract
The present invention provides a unique and novel system for acquiring digital images of identification documents such as drivers license, passports, or medical insurance records using a digital camera so as to establish a high resolution image and extracting data automatically with machine vision tools so as to acquire accurate data. The present invention teaches also a system that acquires multi-spectral images of both sides of the identification document.
31 Citations
22 Claims
-
1. A system for extracting information from an identity document comprising at least one light source for illuminating the identity document;
- at least one digital camera comprising a lens and a two-dimensional sensor array for acquiring at least one two-dimensional image of the identity document; and
a processor for processing said at least one two-dimensional image, said processor configured to run an optical character recognition (OCR) routine for extracting textual information from the identity document in the form of alphanumerical information, wherein said extracting textual information comprises the following steps;a) determining boundaries between text and background based on minimum contrast between the character and other colored data on the identity document; b) extracting black and white glyphs using at least a portion of said determined boundaries; c) applying said OCR routine on said black and white glyphs, thereby recognizing symbols and characters; and d) sending said recognized symbols and characters as alphanumerical information. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19)
- at least one digital camera comprising a lens and a two-dimensional sensor array for acquiring at least one two-dimensional image of the identity document; and
-
20. A method for extracting information from an identity document, comprising the steps of:
-
a) providing a system for extracting information from an identity document including; i. at least one light source for illuminating the identity document; ii. at least one digital camera having a lens and a two-dimensional sensor array for acquiring at least one two-dimensional image of the identity document; and iii. a processor for processing said at least one two-dimensional image; b) determining boundaries between text and background based on minimum contrast between the character and other colored data on said image of the identity document; c) extracting black and white glyphs using at least a portion of said determined boundaries; d) applying an optical character recognition (OCR) routine on said black and white glyphs, thereby recognizing symbols and characters; and e) sending said recognized symbols and characters as alphanumerical information; wherein said determining of the type of the identity document and said identifying said textual field regions in the identity document, are performed before applying said OCR routine on said black and white glyphs. - View Dependent Claims (21, 22)
-
Specification