Computer graphics processing and selective visual display system – Computer graphics processing – Graphic manipulation
Reexamination Certificate
2000-07-26
2004-11-30
Cabeca, John (Department: 2173)
Computer graphics processing and selective visual display system
Computer graphics processing
Graphic manipulation
C345S647000, C345S653000, C345S960000, C382S141000
Reexamination Certificate
active
06825856
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the translation of human visual inspection analysis of an object into physical quantifiable parameters used in an inspection device via a graphical user interface. The present invention further relates to displaying a multidimensional depiction of a manufactured item or classifiable item and then morphing the three dimensional depiction to establish maximum and minimum ranges for various dimensions and aspects of the manufactured item. More particularly, the present invention relates to a method and apparatus for viewing a three dimensional depiction of a solder joint then morphing the depiction of the solder joint to establish maximum and minimum acceptable limits for such solder joint being viewed and tested in a classification/manufacturing test apparatus.
2. Description of Related Art
One goal of inspection processes is to classify a device which is under inspection into one of two (or many) categories based on some criteria (i.e. weight: 3 lbs. or greater is a pass; less than 3 lbs. fails). There are many different types of inspection processes. One type might use a human inspector. Human inspectors are common in visual inspections. Another type of inspection might use a tool, machine or device such as a caliper to measure, for example, an inside diameter of a pipe.
Consider for a moment the contrast between human inspection and tool inspection systems. In a human inspection system the human utilizes human visual and judgment systems. The parameters used to describe or measure an item or object under inspection may include traits such as “smooth”, “shiny”, “crooked”, etc. The traits are fuzzy (not precisely quantified as are physical quantifiable parameters) and are usually distillations (condensations) of a complex (large number of and/or interdependent) set of physical, quantifiable parameters. For example, a trait of a textured surface may be “smoothness”. The surface texture, under very close inspection, has thousands of tiny pits, each having five dimensions (three volumetric dimensions, and two positional dimensions). Regardless of the potentially thousands of physical quantifiable parameters required to define smoothness, human visual inspection can quantify the fuzzy trait of smoothness relatively easily. The measures of human visual inspection tend to be inexact (e.g. somewhat, very, or extremely—smooth).
Conversely, a tool measurement system uses substantially only physical quantifiable parameters which are often standardized measures (and in many instances defined by the National Institute of Standards and Technology). Examples of physical quantifiable parameters are length, weight, temperature, frequency, density, etc. The measures of physically quantifiable parameters are generally numeric values (15, 12.5, 0.0026).
Automated inspection systems have become an important part of many production and item inspection facilities. Quality control and the ability to know when a production line is producing good, marginal or poorly manufactured products with respect to predetermined specifications is paramount in today's industrialized/information based society.
An automated test system may test and inspect manufactured objects, such as solder joints, using cross-sectional x-ray imaging or laminography. Such a system may detect defects in solder joints on printed circuit board assemblies (PCBA's) that are single sided or doubled sided. Two-dimensional x-ray views are taken of the board. The testing and inspection is non-destructive because physical contact with the object of inspection is not required.
Drawbacks of automated test equipment tend to be related to the set-up and the programming of the equipment. For example, a set-up and programming processes for a test/inspection machine may be as follows: First the limits must be set. As such, a programmer obtains a “good” reference “item” (e.g. PCBA) to be tested. Then the programmer utilizes a test/inspection device and obtains the device's reported parameters. The parameters are usually reported as numerals which represent different parameters of an “object”. The programmer then estimates how much the numerical parameters can vary on the object and be within tolerance. Numerical parameter estimation is a determination of physical dimension tolerance limits which are utilized by an inspection machine. The programmer is effectively setting appropriate tolerance limits for the physical dimensions of the object. Commonly, two numbers are estimated and manually input for each parameter by the programmer. The two numbers may be for an upper and lower limit. When an object, such as a solder joint, has many parameters and physical dimensions (e.g. more than 50) and when there are multiple objects (types and subtypes) on an item for inspection (e.g. PCBA), then a programmer may have to make hundreds of estimations in order to set-up and program a test/inspection machine. The programmer must manually enter the estimations into the test/inspection database for the item.
In order to verify the limits estimated and entered by the programmer, the programmer may run a series of items through the test/inspection device. The test/inspection device will measure and extract parameters of objects on the item and compare the extracted measurements against the estimated limits entered by the programmer. Any extracted measurment that is outside the estimated limits will be indicated as “failed”. Any extracted measurement that is inside the estimated limits is indicated as “passed”. The programmer will then visually inspect the objects (solder joints) to determine if the entered estimates are classifying the objects correctly. In many instances the estimates do not correctly classify the objects.
The programmer must adjust the appropriate limits if entered estimates do not classify the object correctly. To do so, the programmer looks at the object's shape to determine what parameter(s) of the object is causing the object to be classified incorrectly. Then the programmer must determine (usually guess based on experience) which estimated limit should be adjusted. Sometimes this is a fairly straight forward process when there is a one-to-one relationship between the parameter and the estimated limit (e.g. width of an object and the number describing the width of the object). Other times, this process of “guessing” is not straight forward when multiple parameters interact (e.g. slope, position, and height of an edge). The programmer now must re-estimate how much to change the value of the estimated limit(s).
This process of setting-up and programming a test/inspection machine has various problems. Such problems include that it is tedious and slow. There are a lot of entries to understand and make. This process is highly subjected to human error, human repeatability errors, and human fatigue errors. This process further has the drawback of requiring the programmer to have a cognitive ability to relate physical parameters (e.g. related to shape) to numerical parameters. This is very difficult when a human must visualize
10
or more parameters simultaneously.
Another drawback is that the programmer must translate the visual form (or deformity) to a parameteric value. Furthermore, due to the programming process being tedious and difficult, many programmers “give up” before the system is well programmed or tuned. Also, even if a system is well tuned, operators may believe that the limits are not set correctly and therefore allow out of spec parts to “pass” inspection.
With laminography a two-dimensional view of, for example, a solder joint can be taken. The two-dimensional data can be turned into a plurality of data files and then manipulated manually, by a system programmer, to set maximum and minimum dimensions or tolerances for the inspection/test device to utilize.
The system programmer takes the standard or ideal dimensions for a manufactured item, such as a solder joint, and by hand manually enters numbers or data to set the maximum, m
Buck Dean C
Fazzio Ronald Shane
Agilent Technologie,s Inc.
Cabeca John
Pillai Namitha
LandOfFree
Method and apparatus for extracting measurement information... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for extracting measurement information..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for extracting measurement information... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3287453