Image analysis – Applications – Manufacturing or product inspection
Reexamination Certificate
1997-12-16
2001-01-09
Bella, Matthew C. (Department: 2721)
Image analysis
Applications
Manufacturing or product inspection
C382S145000, C382S148000, C348S126000, 36, 36, C250S559040
Reexamination Certificate
active
06173071
ABSTRACT:
This invention relates generally to the manufacture of printed circuit boards, and more specifically, to manufacturing defect analyzers.
Printed circuit boards (PCB's) are typically inspected during a manufacturing process for determining whether the boards contain manufacturing defects. As a result, defective PCBs can be identified before they are incorporated into electronic products, thereby minimizing the chance that the electronic products will fail prematurely in the field.
One method of inspecting PCBs is known as automatic optical inspection (AOI), which generally uses an illumination device and video cameras mounted in an inspection head. A test computer typically controls movement of the inspection head relative to a board under inspection (BUI), thereby allowing the cameras to scan a surface of the BUI and capture video data representative of the scanned surface. The captured video data is then stored for subsequent processing. This method is used in INTERSCAN™ Automated Optical Inspection Systems, sold by TERADYNE®, Inc., Walnut Creek, Calif., USA.
The test computer typically generates a list of windows where particular features of the BUI or components attached to the BUI are to be inspected. Each window generally indicates a rectangular region on the BUI.
The video cameras then typically scan each window for capturing video data representative of the region of the BUI encompassed by the window. The captured video data generally includes information relating to pixels corresponding to each encompassed region of the BUI. The captured video data is then typically stored in a frame storage unit.
Next, the test computer typically processes the captured video data by executing algorithms. One commonly executed algorithm determines an average level of brightness for each window. This algorithm, for example, can be used for detecting solder bridges, which affect the reflectivity of the BUI and therefore the brightness level of corresponding windows.
In general, each pixel is represented by a byte of captured video data, and the value of the byte of video data is proportional to the brightness level. Accordingly, the algorithm for determining the average brightness level of a window can be described by the following pseudo-code, in which “LEFT_EDGE,” “RIGHT_EDGE,” “TOP_EDGE,” and “BOTTOM_EDGE” refer to the boundaries of a rectangular window, and “X” and “Y” define a position for each pixel within the window:
X1=LEFT_EDGE;
X2=RIGHT_EDGE;
Y1=TOP_EDGE;
Y2=BOTTOM_EDGE;
SUM=0;
FOR X=X1 TO X2 STEP_SIZE 1
FOR Y=Y1 TO Y2 STEP_SIZE 1
GET PIXEL(X,Y);
SUM=SUM+PIXEL(X,Y);
NEXT Y;
NEXT X;
AVERAGE=SUM/[(X2−X1)*(Y2−Y1)];
In particular, the line of pseudo-code, “GET PIXEL (X,Y),” is meant to describe the step of reading information relating to the pixel at position (X,Y) from the frame storage unit. Further, the line, “SUM=SUM+PIXEL(X,Y),” is meant to describe the step of summing the value of the pixel at position (X,Y) with the values of other pixels within the window. Finally, the line, “AVERAGE=SUM/(X2−X1)*(Y2−Y1),” is meant to describe the step of dividing the “SUM” by the total number of pixels within the window, for obtaining the average value of the pixels or the average brightness level of the window.
However, we have recognized that analyzing captured video data in this way has shortcomings. In particular, the steps of reading pixel information and summing pixel values generally must be repeated for every pixel within each window. This requires a significant amount of processing time.
Although manufacturing defect analyzers may have system busses that are wide enough for reading information relating to several adjacent pixels at one time, the mere act of reading information relating to more than one pixel does not always reduce the overall processing time. This is because masking and shifting of the pixel information is generally required to ensure that each bit of pixel data has the proper significance during the summing step. The required acts of masking and shifting of data also contribute significantly to processing time.
It would therefore be desirable to have a manufacturing defect analyzer that can inspect PCB's at a rate that is faster than current methods. Such a defect analyzer would still be able to inspect the PCB's using traditional inspection algorithms.
SUMMARY OF THE INVENTION
With the foregoing background in mind, it is an object of the invention to provide a manufacturing defect analyzer that reliably detects defects on printed circuit boards at a faster rate.
The following and other objects are achieved by an apparatus for inspecting a printed circuit board including a test computer, an improved bus interface, an inspection head, and at least one frame storage unit. The improved bus interface includes an input coupled to the frame storage unit, an output coupled to the test computer, circuitry for adding values of digital data coupled to the input, and circuitry for selecting between the adding circuitry output and the input. The digital data is provided by the frame storage unit. Further, the selecting circuitry is controlled by the test computer.
Still further objects and advantages will become apparent from a consideration of the ensuing description and drawings.
REFERENCES:
patent: 4185298 (1980-01-01), Billet et al.
patent: 4395697 (1983-07-01), Dargel et al.
patent: 4541114 (1985-09-01), Rutenbar et al.
patent: 4710923 (1987-12-01), Nagumo
patent: 5103105 (1992-04-01), Ikegaya et al.
patent: 5153726 (1992-10-01), Billing
patent: 5245421 (1993-09-01), Robertson et al.
patent: 5260779 (1993-11-01), Wasserman
patent: 5282268 (1994-01-01), Mieras et al.
patent: 5319584 (1994-06-01), Ooyabu
patent: 5325215 (1994-06-01), Shibata et al.
patent: 5495337 (1996-02-01), Goshorn et al.
patent: 5598345 (1997-01-01), Tokura
Fuhriman Gregg E.
Wasserman Harold
Bella Matthew C.
Chawan Sheela
Teradyne Legal Department
LandOfFree
Apparatus and method for processing video data in automatic... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Apparatus and method for processing video data in automatic..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Apparatus and method for processing video data in automatic... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2473833