Optics: measuring and testing – Dimension – Width or diameter
Reexamination Certificate
2002-07-02
2004-11-30
Rosenberger, Richard A. (Department: 2877)
Optics: measuring and testing
Dimension
Width or diameter
C356S625000
Reexamination Certificate
active
06825939
ABSTRACT:
CROSS REFERENCE TO RELATED APPLICATIONS
This application claims priority of the German patent application 101 31 897.9 which is incorporated by reference herein.
FIELD OF THE INVENTION
The present invention relates to a method and a measurement apparatus for detection of a specimen, a specimen being illuminated with a light source and imaged with the aid of an imaging optical system onto a detector preferably embodied as a CCD camera, and the specimen being detected repeatedly with the detector.
BACKGROUND OF THE INVENTION
Methods for the detection of specimens are known from practical use. In industrial applications, for example in the metrology of line widths or positions on substrates of the semiconductor industry, coordinate measuring instruments such as those described, for example, in German Patent Application DE 198 19 492.7-52 are used. This measuring instrument serves for highly accurate measurement of the coordinates of features on substrates, e.g. masks, wafers, flat screens, and evaporatively deposited features, but in particular for transparent substrates. The coordinates are determined relative to a reference point, to an accuracy of a few nanometers. In this context, for example, a specimen is illuminated with light of a mercury vapor lamp and is imaged by means of the imaging optical system onto a CCD camera. The CCD camera usually acquires several images of the same specimen using the same exposure time. The images thus obtained are conveyed to a statistical analysis operation. The statistical analysis operation comprises, for example, an averaging of the detected image data. The overall result thus obtained has better repeatability as compared to analysis of an individual image.
Methods for the detection of a specimen in which the specimen is detected repeatedly with a detector can be problematic if, for example, changes in illumination intensity occur during the period of the individual detections. In the case of the coordinate measuring instrument, the specimens detected are principally ones that have defined specimen features, the specimen features in turn having linear edges. One of the tasks of a coordinate measuring instrument is to determine the location of the edges of the specimens as accurately as possible relative to a reference point. After the detection of several images of the same specimen with a CCD camera, it has been found that in the event of even slight fluctuations in the exposure time of the CCD camera, the edge location after a statistical analysis of the detected image data has changed. Attempts have accordingly been made to select the target range of the automatic exposure system as narrowly as possible, so that the location of the edges in the detected image data does not change. This procedure is not possible for all regions of the same specimen, however, so that a certain residual error in edge location has hitherto been unavoidable. In addition, the change in the detection of the edge location can also be based on residual errors that are not directly associated with the exposure time of the CCD camera. These residual errors can be of systematic or random origin. Ultimately, what is desirable is to minimize all the residual errors and the changes in data analysis associated with them.
SUMMARY OF THE INVENTION
It is therefore the object of the present invention to describe and develop a method for detection of a specimen in which the fluctuations in the statistical analysis of detected signals or data are minimized, the detected signals or data being subject to detection-related error sources.
A further object of the present invention is to describe and develop a measurement apparatus for detection of a specimen in which the fluctuations in the statistical analysis of detected signals or data can be minimized, the detected signals or data being subject to detection-related error sources.
A method for detection of a specimen according to the present invention comprises the steps:
illuminating a specimen with a light source,
imaging said specimen with the aid of an imaging optical system onto a detector,
detecting said specimen repeatedly with the detector,
and thereby varying the detection time of the detector for the individual detections or the intensity of the light serving for specimen illumination.
Another method for detection of a specimen according to the present invention comprises the steps:
illuminating a specimen with a light source,
imaging said specimen with the aid of an imaging optical system onto a detector,
detecting said specimen repeatedly with the detector,
and thereby varying the detection time of the detector for the individual detections and the intensity of the light serving for specimen illumination.
What has been recognized according to the present invention is firstly that external influences such as, for example, slight fluctuations in the light intensity of the light source can be prevented, if at all, only with a great deal of effort. Even optimization of the automatic detection time system, with the goal of setting the optimum detector detection time for each specimen detection, cannot yield error-free results in every case. If additional random or systematic specimen detection errors are also to be minimized, then according to the present invention once again no analysis is made of the cause underlying the individual errors, in order then to minimize each of them with a corresponding error-reducing action.
Instead, according to the present invention, the detection time of the detector is varied for each individual detection. Varying the detection time firstly increases the noise bandwidth of the measurement results, since the influences of the different detection times are incorporated into the result. The statistical analysis of the measured data, however, averages over all the changeable error components. In this respect, therefore, only the average error due to detection time influences in the elapsed detection time interval is incorporated into the measurement result. At a constant detection time, however, the entire or constant error component is incorporated into the detection, which becomes apparent in the context of a further detection or a further detection series with modified detection conditions.
According to the present invention, therefore, no provision is made for further development of the automatic detection time system with the intention of achieving the same detection result in changed environmental conditions; instead, the detection time of the detector is deliberately varied so that subsequent statistical analysis of the detected measured data (largely independently of the changed environment conditions) yields a result that is very close to the actual expected result, minimizing the error due to detection time variation.
Additionally or alternatively, according to the present invention provision is made for the intensity of the light serving for specimen illumination to be varied for each individual detection. This could be accomplished, for example, with a rotating neutral density filter disk arranged in the illumination beam path.
The variation in detection time and/or the variation in the intensity of the light serving for specimen illumination must, however, be accomplished in such a way that the detected signals lie in the range of permissible detector modulation, i.e. they must not exceed the dynamic range of the detector. In a corresponding method step, provision is accordingly made for determining the limits of the permissible detector modulation in the context of a calibration operation. A variation in detection time is then advisable only within those limits. For some applications, however, provision could also be made for deliberate overmodulation during a detection, for example in order to be able to depict specimen regions with different intensities in the low dynamic range of the detector, while other specimen regions yield intensities that are detected in overmodulated fashion with that setting for the detection time.
The variation in detection time and/or in the
Barth Vincent P.
Leica Microsystems Semiconductor GmbH
Rosenberger Richard A.
LandOfFree
Method and measuring arrangement for detecting an object 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 measuring arrangement for detecting an object, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and measuring arrangement for detecting an object will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3334749