Browse Prior Art Database

Registration Marks for Machine Vision

IP.com Disclosure Number: IPCOM000104741D
Original Publication Date: 1993-May-01
Included in the Prior Art Database: 2005-Mar-19
Document File: 1 page(s) / 32K

Publishing Venue

IBM

Related People

Lytle Jr, J: AUTHOR [+2]

Abstract

Machine vision problems often involve the location of objects. In industrial machine vision applications, objects must be located precisely prior to assembly, inspection, or character recognition tasks may be performed. In document imaging applications, it is necessary to locate regions of interest within an image prior to performing Optical Character Recognition (OCR). The registration marks for machine vision disclosed here were developed to locate an address label on tax return images; however, the concept is applicable to the full spectrum of imaging and industrial machine vision applications.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 100% of the total text.

Registration Marks for Machine Vision

      Machine vision problems often involve the location of objects.
In industrial machine vision applications, objects must be located
precisely prior to assembly, inspection, or character recognition
tasks may be performed.  In document imaging applications, it is
necessary to locate regions of interest within an image prior to
performing Optical Character Recognition (OCR).  The registration
marks for machine vision disclosed here were developed to locate an
address label on tax return images; however, the concept is
applicable to the full spectrum of imaging and industrial machine
vision applications.

      The figure shows a tax return label with registration marks
added.  Three marks determine the object (label) orientation without
ambiguity.  The circular dots are intended to provide a course
measurement of the object's position.  Circular dots are rotationally
symmetrical and slightly larger than other objects (characters) in
the image.  Thus, the dots are easily located using correlation or
histogram search techniques.  The adjacent line segments are included
to provide a more precise position and an angular measure.

Disclosed Anonymously.