Browse Prior Art Database

Algorithm for Detecting Solder Joint Defects using K-L Transformation and Fourier Transformation

IP.com Disclosure Number: IPCOM000115105D
Original Publication Date: 1995-Mar-01
Included in the Prior Art Database: 2005-Mar-30
Document File: 6 page(s) / 201K

Publishing Venue

IBM

Related People

Fujita, T: AUTHOR [+3]

Abstract

This article describes an algorithm that would be effective in detecting defects of solder joints for fine pitch Quad Flat Pack (QFP) components using video images. First of all, several characteristic values (brightness of each joints) will be obtained from video images. The main mathematical treatment is to make linear combination of characteristic values being referred to as the value P. If P is greater than Pc (threshold value), this joint will be judged as bad joint. Characteristic values will be K-L transformed so as to calculate weighting factors of linear combination. This operation is carried out before the current inspection process. The threshold value Pc will be determined by carrying out statistical treatment of teaching card data.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 35% of the total text.

Algorithm for Detecting Solder Joint Defects using K-L Transformation
and Fourier Transformation

      This article describes an algorithm that would be effective in
detecting defects of solder joints for fine pitch Quad Flat Pack
(QFP) components using video images.  First of all, several
characteristic values (brightness of each joints) will be obtained
from video images.  The main mathematical treatment is to make linear
combination of characteristic values being referred to as the value
P.  If P is greater than Pc (threshold value), this joint will be
judged as bad joint.  Characteristic values will be K-L transformed
so as to calculate weighting factors of linear combination.  This
operation is carried out before the current inspection process.  The
threshold value Pc will be determined by carrying out statistical
treatment of teaching card data.  The results of judgement were well
consistent with the results of inspection obtained by inspector.

      Many methods for classifying something, having N data a1-aN,
into two groups, i.e., group A and group B have been developed so
far.  In soldering inspection machine, each joints being accompanied
by N data are judged as good or bad based on these data(a1-aN).  An
example of these methods is described in the Table according with the
necessity of teaching data.  These methods are defined as the way for
establishing the black box in the manner that each input data are
well classified as we expected.  Using experimental data obtained
artificially, the mechanism of the black box can be established.
                              Table 1
        Method (concept)                        Teacher
    Discrimination Analysis               Need teacher (discrete)
    Multivariate analysis                 Need teacher (continuous)
    Neural Net-work Analysis              Need teacher (discrete)
    Principal Component Analysis          No teacher
    (K-L transformation)

      Among these various methods, the Principal Component Analysis
(PCA) was chosen as being suitable for the algorithm for the
detection of detect joints because of no necessity of teacher data.
PCA is often referred to as K-L transformation in the field of
images.  The same mathematical treatment is called as PCA in the
field of statistics.  Building up the black box is equivalent to make
an algorithm.  Many data (more than two) will be input into the black
box where a kind of calculation will be performed, then output data,
on which judgement is based, will be generated.

      Based on this output data, each joint will be classified into
two groups, one for good joint and the other for bad joint.  The
content of black box being developed is well explained in Fig. 1.

      The mechanism of this black box is somewhat very simple.  At
first, input data will be linearly tr...