Browse Prior Art Database

Automated Calibration of Card Inspection Tool

IP.com Disclosure Number: IPCOM000040252D
Original Publication Date: 1987-Oct-01
Included in the Prior Art Database: 2005-Feb-02
Document File: 2 page(s) / 14K

Publishing Venue

IBM

Related People

Kishi, GT: AUTHOR [+2]

Abstract

Automated calibration is described for a video inspection tool that automatically inspects printed circuit cards to verify correct component lead insertion prior to soldering the card. The tool uses fast, low computation analysis routines for lead verification. These fast routines allow a slow, inexpensive, general-purpose computer to perform the analysis at production speeds. The analysis routines verify lead presence by applying a set of predetermined parameters to the image data. The values of these parameters are dependent upon factors such as clinch angle, lead shape and reflectance, and lighting intensity. The values chosen for these parameters are critical to the inspection results because if they are too "conservative", then many leads will be falsely reported as being defective.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 1 of 2

Automated Calibration of Card Inspection Tool

Automated calibration is described for a video inspection tool that automatically inspects printed circuit cards to verify correct component lead insertion prior to soldering the card. The tool uses fast, low computation analysis routines for lead verification. These fast routines allow a slow, inexpensive, general-purpose computer to perform the analysis at production speeds. The analysis routines verify lead presence by applying a set of predetermined parameters to the image data. The values of these parameters are dependent upon factors such as clinch angle, lead shape and reflectance, and lighting intensity. The values chosen for these parameters are critical to the inspection results because if they are too "conservative", then many leads will be falsely reported as being defective. On the other hand, if the values are too "liberal," then some defective leads may be accepted as being good. The reason for the false rejects or the false accepts is because the video images for any given lead type varies considerably on any card. There is no set of parameter values that will always give perfect results (neither false accepts nor false rejects.) The values chosen are a compromise which hold false rejects to an acceptable level while virtually eliminating false accepts. In addition, on production jobs there is no single set of parameter values used by the analysis routines that will work on all cards. In fact, different jobs of the same part number may require different parameter values. Therefore, these parameter values may require adjustment every time a new job is run on the machine. The auto-calibration routine determines what the best compromise values are for a new job. Using the first carrier (a fixture that holds four cards) of the job, selected grayscale pixel intensities from the holes and different lead type images are measured. The distributions of these grayscale intensities are used to determine the best set of compromise values to use to inspect the rest of the job. Based on the distributions, the auto-calibration routine picks grayscale thresholds that will be used by the analysi...