Image analysis – Applications – Manufacturing or product inspection
Reexamination Certificate
2000-07-07
2001-08-21
Mehta, Bhavesh (Department: 2621)
Image analysis
Applications
Manufacturing or product inspection
C382S141000
Reexamination Certificate
active
06278796
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Reservation of Copyright
The disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the U.S. Patent and Trademark Office patent files or records, but otherwise reserves all copyrights whatsoever.
2. Field of Invention
The present invention relates to image processing, and, more particularly, to a method and apparatus for subsampling an image so that the subsampled image may be further processed within certain constraints.
3. Description of Background Information
In the field of image processing, whatever its ultimate application, the general conventional approach is that shown in
FIG. 1
, wherein first an image is obtained (at P
10
) and then the image is processed (at P
12
). Typically, the processing of the image becomes more accurate as the resolution of the image increases. That is to say, the greater the resolution of the image, the more accurate the result of the image processing. However, image processing time is almost always also a function of the resolution of the image, so that the greater the resolution (and therefore the accuracy), the greater the amount of time taken to process the image.
If image processing time is not an issue, the best possible accuracy can be achieved. However, in many applications, processing time and memory space constraints exist, and consequently accuracy must be sacrificed to operate, if possible, within these constraints.
In general, performing image processing (P
12
in FIG.
1
), using the entire input image may cause the processing to take too long and/or require the use of too much memory space. On the other hand, using too little of the input image may cause the result of subsequent image processing to not be sufficiently accurate.
SUMMARY OF THE INVENTION
The present invention is provided to improve upon machine vision and image processing systems. The present invention may be further provided to allow the image processing phase of an image processing system to run within a constraint satisfying time range while simultaneously achieving a result within a required accuracy range. The present invention may be still further provided to allow an image processing algorithm in an image processing system to run within an optimal time range while simultaneously achieving a result within a required accuracy range.
In order to achieve these ends, one or more aspects of the present invention may be followed in order to bring about one or more specific objects and advantages, such as those noted below.
One object of the present invention is to provide an image processing system which subsamples, in a constraint satisfying manner, an input image based on the size of the image and on the expected time, space and accuracy requirements of subsequent (downstream), and possibly prior (upstream), computational processes.
Another object of this invention is to provide an image processing system which subsamples, in a constraint satisfying manner, an input image based on the size of the image and on the expected time, space and accuracy requirements of other computational processes. The other processes may be prior to the subsampling, after the subsampling, or both. The requirements may also include implementation limitations, such as, for example, allowed image sizes and allowed subsampling factor(s) and the like.
One aspect this invention is a method, in an image processing system, of subsampling, in a constraint satisfying manner, an image for subsequent processing of the subsampled image. The method includes providing constraints for the prior and/or subsequent image processing; and subsampling the image based on at least the image size so as to satisfy the provided constraints.
In another aspect, this invention is a computer-readable medium encoded with a program for subsampling, in a constraint satisfying manner, an image for subsequent processing of the subsampled image. The program includes providing constraints for the prior and/or subsequent image processing; and subsampling the image based on at least the image size so as to satisfy the provided constraints.
In yet another aspect, this invention is a device, in an image processing system, for subsampling, in a constraint satisfying manner, an image for subsequent processing of the subsampled image. The device includes means for providing constraints for the prior and/or subsequent image processing; and means for subsampling the image based on at least the image size so as to satisfy the provided constraints.
In some preferred embodiments, the subsampling means comprises means for determining a subsampling factor (or factors) for the image; and means for subsampling the image by the subsampling factor(s).
Preferably the subsampling includes determining a subsampling factor or factors for the image; and subsampling the image by the subsampling factor or factors.
In preferred embodiments the constraints are performance constraints.
In preferred embodiments the constraints comprise at least one of time constraints, uncertainty constraints and accuracy constraints. In some situations the uncertainty constraints are the inverse of the accuracy constraints. Time constraints can include an estimate of the time needed for both the subsampling and the subsequent image processing as well as, possibly, the time needed for prior processing. This estimate is expressed as a function of either the size of the subsampled image (width and height) or of the subsampling factors. The time constraints also include at least one of a maximum and a minimum time that both the subsampling and the subsequent (and possibly prior) image processing should take. Uncertainty constraints include an estimate of the uncertainty in one or more results of subsequent image processing. This uncertainty estimate is a function of either the size of the subsampled image (width and height) or of the subsampling factors. The uncertainty constraints also include at least one of a maximum and minimum uncertainty required of subsequent image processing.
Accuracy constraints include an estimate of the accuracy in one or more results of subsequent image processing. This accuracy estimate is a function of either the size of the subsampled image (width and height) or of the subsampling factors. The accuracy constraints also include at least one of a maximum or minimum accuracy required of subsequent image processing.
The constraints also include implementation (that is, hardware/software) related constraints such as, for example, constraints relating to allowed image sizes, allowed subsampling factors and the like.
REFERENCES:
patent: 4728195 (1988-03-01), Silver
patent: 4752898 (1988-06-01), Koenig
patent: 4972359 (1990-11-01), Silver et al.
patent: 5179441 (1993-01-01), Anderson et al.
patent: 5255352 (1993-10-01), Falk
patent: 5291563 (1994-03-01), Maeda
patent: 5367439 (1994-11-01), Mayer et al.
patent: 5371690 (1994-12-01), Engel et al.
patent: 5511137 (1996-04-01), Okada
patent: 5513275 (1996-04-01), Khalaj et al.
patent: 5517235 (1996-05-01), Wasserman
patent: 5694486 (1997-12-01), Shigeeda et al.
patent: 5696848 (1997-12-01), Patti et al.
patent: 5724439 (1998-03-01), Mizuoka et al.
patent: 5758043 (1998-05-01), Takizawa et al.
patent: 5764536 (1998-06-01), Yamamoto et al.
patent: 5768409 (1998-06-01), Csipkes et al.
patent: 5784484 (1998-07-01), Umezawa
patent: 5995648 (1999-11-01), Drisko et al.
patent: 6157732 (2000-12-01), Drisko et al.
Cognex 3000/4000/5000 Vision Tools Revision #7.4 #590.0136 Users Manual, No Date.
Cognex 4000/5000 SMD Placement Guide Package Release 3.2 #590.1039 User Manual, No. Date.
Cognex 4000/5000 SMD Placement Guidance Package User's Manual , rev. 3.2, 1996, Ch. 7, pp. 269-310.
Cognex 4000-5000 SMD Placement Guidance Package User's Manual, rev. 3.2, 1996, Ch. 11, pp. 357-372.
Cognex 3000/4000/5000 Programmable Vision Engines Image Processing, rev. 7.4, 1996, Ch. 2, pp.
Bachelder Ivan A.
Drisko Robert
Cognex Corporation
Irving Richard C.
Mehta Bhavesh
Weinzimmer Russ
LandOfFree
Image processing system and method using subsampling with... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image processing system and method using subsampling with..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing system and method using subsampling with... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2452225