Browse Prior Art Database

Diagonal Signature

IP.com Disclosure Number: IPCOM000103421D
Original Publication Date: 1990-Nov-01
Included in the Prior Art Database: 2005-Mar-17
Document File: 1 page(s) / 45K

Publishing Venue

IBM

Related People

Beaumont, GD: AUTHOR [+2]

Abstract

This invention is a technique for locating objects in a displayed image by exploiting vertical arithmetic hardware available in a frame grabber card in a video processing system. The technique may be used in a manufacturing concept of machine vision where continuous testing adjustments are made to counteract machine imprecisions and part distortions.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 82% of the total text.

Diagonal Signature

      This invention is a technique for locating objects in a
displayed image by exploiting vertical arithmetic hardware available
in a frame grabber card in a video processing system.  The technique
may be used in a manufacturing concept of machine vision where
continuous testing adjustments are made to counteract machine
imprecisions and part distortions.

      The initial step of object location is to determine the
signature or projection of the object on the "X" axis.  The last line
at the bottom of the screen is the profile of the intensity of the
sum of all pixels of the object for each particular column.  This
profile is used to identify the middle of the object on the "X" axis.

      Where the hardware for the video processing does not support
the projection of the object on the "Y" axis, the middle of the
object on the "Y" axis may be determined by a method called "diagonal
signature".  After the middle of the object on the "X" axis is
determined as above, the hardware is then adjusted to shift each
subsequent row of two additional pixels to the left for each row on
the screen.  A signature on the "X" axis is again performed.  This
activity will produce a diagonal signature of the picture.  The pixel
intensity on the left side and the bottom of the screen is read.  The
bottom of the screen will be a summation of the intensities of the
pixels for each column and the left side will be a partial sum of the
first column.  A vec...