Browse Prior Art Database

Method for Automatically Comparing a Printable Output to a Digitized Input

IP.com Disclosure Number: IPCOM000121866D
Original Publication Date: 1991-Oct-01
Included in the Prior Art Database: 2005-Apr-03
Document File: 2 page(s) / 74K

Publishing Venue

IBM

Related People

Feenstra, DD: AUTHOR [+5]

Abstract

Described is a new and unique way of using existing tools. A general description follows: - The images have to be in some form of raster scan graphics format. The two used for automatic label generation are PSEG38PP and BIT MAP. - The source image dataset is read into the graphic editor (in our case YIMAGE) and saved into a temporary workspace. - The target image dataset is read into the graphic editor. A reverse video operation is performed to complement the PELs from light to dark and dark to light. This reversed image is saved to a temporary workspace. - The two temporary workspaces are logically ORed to get a sum of all the dark (or light, depending on your reference point) pels. If the two images match identically, you should receive a BIT MAP where all pels are light or dark.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Method for Automatically Comparing a Printable Output to a Digitized
Input

      Described is a new and unique way of using existing
tools. A general description follows:
-  The images have to be in some form of raster scan graphics format.
The two used for automatic label generation are PSEG38PP and BIT MAP.
-  The source image dataset is read into the graphic editor (in our
case YIMAGE) and saved into a temporary workspace.
-  The target image dataset is read into the graphic editor. A
reverse video operation is performed to complement the PELs from
light to dark and dark to light.  This reversed image is saved to a
temporary workspace.
-  The two temporary workspaces are logically ORed to get a sum of
all the dark (or light, depending on your reference point) pels. If
the two images match identically, you should receive a BIT MAP where
all pels are light or dark. If there are some that are the opposite,
these represent the areas where the source and target do not match.
In our specific application, this was done with an APL matrix
operation.
-  To determine goodness, the tolerance of the image (as defined in
the original CADAM or CATIA drawing) must be converted to pels using
the pel resolution number of the graphic format being used (in our
case 240 pels per inch in both the vertical and horizontal
direction).  This will give the number of PELs in a row that can be
light (or dark depending on the reference point again).
-  Each row will be scanned and each column to make sure that the
miscompare areas do not exceed the tolerance calculated in the last
section.
-  This process will be repeated doing the reverse video operation on
the source instead of the target and recomparing.  This second step
is needed in case one part of one of the images completely overlays
that part of a second image.

      T...