Browse Prior Art Database

Composite Image Integration to Facilitate Vision Teaching

IP.com Disclosure Number: IPCOM000123315D
Original Publication Date: 1998-Sep-01
Included in the Prior Art Database: 2005-Apr-04
Document File: 1 page(s) / 42K

Publishing Venue

IBM

Related People

Menard, EB: AUTHOR [+2]

Abstract

Disclosed is a technique to improve the teaching process for Automated Optical Inspection (AOI) equipment. The technique requires processing a number of samples of the objects to be inspected, for example, assembled circuit boards. The image for each sample are captured and digitized, as usual. For large objects, multiple image windows are generally used. This technique is applied to each image window.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 72% of the total text.

Composite Image Integration to Facilitate Vision Teaching

   Disclosed is a technique to improve the teaching process
for Automated Optical Inspection (AOI) equipment.  The technique
requires processing a number of samples of the objects to be
inspected, for example, assembled circuit boards.  The image for
each sample are captured and digitized, as usual.  For large
objects, multiple image windows are generally used.  This technique
is applied to each image window.

   As the images are taken, a composite average image is
constructed in the machine's memory by averaging each pixel location
across each sample image.  For example, in an 8 bit architecture,
each pixel value would be between 0-255, representing a gray level of
that small portion of the circuit board.  If 3 samples were processed
with values of 110, 116, and 98, the average value is 108.

   Each pixel is averaged with its corresponding pixels across
all of the samples taken.  The resulting set of pixel averages
represents the mean of the samples of EACH pixel.  This averaging
can be updated after each sample is digitized, by appropriately
weighting the current average by the number of samples

   This set of pixels can be displayed as a composite image
having the appearance of an average of all the samples.  This
composite image is presented to the human programmer to facilitate
the manual training process.  The normal process of window setting
and algorithm value setting is performed on this com...