Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Image Processing for Recognition of Circuit Panel Identification Numbers

IP.com Disclosure Number: IPCOM000119920D
Original Publication Date: 1991-Mar-01
Included in the Prior Art Database: 2005-Apr-02
Document File: 2 page(s) / 45K

Publishing Venue

IBM

Related People

Kiyono, S: AUTHOR [+2]

Abstract

Disclosed is an image processing method for recognizing identification numbers represented by drilled holes in printed circuit boards or panels.

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

Image Processing for Recognition of Circuit Panel Identification
Numbers

      Disclosed is an image processing method for recognizing
identification numbers represented by drilled holes in printed
circuit boards or panels.

      The identification (ID) number is represented by drilled holes.
The ID number area is usually located at one edge of the panel.  On
the other hand, the panel may be in a front-side-up or back-side-up
condition while being transferred through the manufactuing line.
When the panel is in the back-side-up condition, the ID number will
be reversed.  One method to deal with such a situation will be to use
two cameras so as to get images of both sides of the panel
simultaneously and to analyze the images to select either one.
However, this method is very time-consuming.

      To avoid the problem, in the proposed scheme, two cameras 1 and
2 are disposd on the same side, as shown in Fig. 1 and the image
processing for ID code recognition is performed in the following way:
      Step 1 : Get image with camera 1.
      Step 2 : Get image with camera 2. Then reverse or rotate the
camera 2 image with respect to an axis which is orthogonal to the
panel transfer direction.
      Step 3 : Fuse the camera 1 image and the reversed camera 2
image.
      Step 4 : Recognize ID number base upon the fused image.

      Fig. 2 illustrates the operation of the proposed image
processing.