Image analysis – Applications – 3-d or stereo imaging analysis
Reexamination Certificate
2000-12-11
2002-07-23
Mehta, Bhavesh (Department: 2621)
Image analysis
Applications
3-d or stereo imaging analysis
C356S003000
Reexamination Certificate
active
06424735
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to high precision cameras of the type which are used to capture a video image of an object with enough precision to allow information to be derived from the video image sufficient to provide accurate measurements of the object for purposes of quality control or other applications requiring accurate dimensional information.
2. Description of Related Art Including Information Disclosed Under 37 CFR 1.97 and 37 CFR 1.98
Even the human eye with its remarkable ability to capture huge amounts of optical information cannot make measurements with precision and accuracy sufficient for quality control of parts requiring highly precise tolerances. Failure to achieve proper measurements can result in a wide range of problems. For example, failure to maintain dimensional tolerances for parts in an automobile engine may result in problems ranging from decreased system life to increase probability of failure under out of the ordinary conditions. Even where later system testing at a manufacturing facility can detect problems resulting from failure to meet dimensional tolerances, unacceptably high rejection rates on account of defective systems can result.
Moreover, even when human eye is capable of performing a required inspection, such factors as fatigue, lighting, distraction and so forth make human inspection unreliable.
Today, in an attempt to minimize those applications where human optical inspection is employed, to avoid the sometimes unpredictable problems caused by fatigue, distraction and other factors, industry has turned increasingly toward the implementation of computerized inspection systems. With such systems, however, the resolution of the optical inspection apparatus is far below that of the human eye.
Currently, many electronic imaging cameras use two dimensional arrays of light sensitive elements, sometimes of the type known as charged coupled devices (CCD) as photodetectors. The purpose of these devices is to convert an optical image into a video image. There are many relatively low priced color and black and white array CCD detectors available for video imaging, but they produce low quality images, as alluded to above. More particularly, CCD detectors and other video imaging devices suffer from a relatively low pixel count. For example, CCD photodetector arrays have the ability to produce quality images with a resolution of approximately 2048 by 2048 pixels. However, these arrays are presently very expensive. Moreover, these very large arrays tend to have defects such as inoperative pixels, inoperative clusters of pixels, or inoperative lines of pixels. When very high quality images are required this type of electronic imaging system is not only very expensive. Such systems may not even be capable of performing a high-quality, high precision measurement.
Linear photodetectors cost much less than array detectors because they have far fewer pixels and thus have correspondingly much higher manufacturing yields. Linear photodetectors, obviously, however, are capable of imaging only one line of information in an image that any given point in time using the single line of photo sensitive devices which they have. Accordingly, the linear photodetector must therefore scan the entire image, line by line.
The same is achieved by using a mechanical scanning assembly for moving the linear photodetector across the image plane in the camera. Generally, systems of this type derived image data by 1) relying upon the precision of the steady state operation of the mechanical scanning assembly, 2) assuming identical transient translational movements during the initiation of a scan, and 3) assuming that translational movement is uniform over time. Such a system, while suitable for making high quality digital images for commercial photography, is inappropriate for use in making high precision measurements.
More particularly, mechanical irregularities in the scanning assembly make the generation of highly precise image information impossible, thus making the image data captured by such systems on suitable for the purpose of confirming dimensional tolerances in a precision manufacturing environment.
Accordingly, it would be advantageous to have a device that will capture a single image of a part with enough precision that accurate measurements can be made of the features of such a part.
SUMMARY OF THE INVENTION
The invention, as claimed, is intended to provide a remedy. It solves the problem of how to provide a linear array-based camera with the capacity of capturing a three dimensional image of a part with enough precision to allow accurate measurements of part features to be made. Moreover, the same is achieved using a positional encoding technique highly insensitive to transient and steady state mechanical tolerances in the mechanical scanning system.
In accordance with the present invention, a positional encoder is coupled to a scanning line array detector. The scanning mechanism moves the linear array along the image plane of a lens. As the linear array moves, each time a predetermined number of resolution steps, corresponding to a resolved distance, are measured by the encoder, the linear array is directed to acquire a line image. In this way, an array of line images, which are precisely spaced, are generated, allowing the precise construction of the entire image on one focal plane. The linear array and focusing optics can then be moved vertically to a different focal plane of the object. The vertical movement of the assembly is in predetermined increments, and also measured by a positional encoder. The scanning process is then repeated on the second focal plane. After repeating the process for all required focal planes, the entire three dimensional image may then be sent to a computer which, using known techniques for detecting object boundaries from digital images, determines the three dimensional position and configuration of the features and compares them to the standard, determining whether the same are within specified tolerances. The focusing optics may be telecentric which may provide for better gauging performance.
Alternatively, in accordance with the present invention the linear array can remain stationary while the three dimensional image of the object is captured. This is accomplished by placing the object on a stage in a known position coupled to a horizontal positional encoder, then horizontally moving the stage through the image capture area of the linear array. As the stage moves, the computer will signal the linear array to send an image back to the computer at a predetermined increment. The linear array and focusing optics can then be vertically moved to a predetermined vertical position corresponding to a new focal plane of the image. Then the horizontal imaging process can be repeated for each new focal plane of the image.
After repeating the process for all required focal planes, the entire three dimensional image may then be sent to a computer which, using known techniques for detecting object boundaries from digital images, determines the dimensional position and configuration of the features and compares them to the standard, determining whether the same are within specified tolerances.
Alternatively, in accordance with present invention the linear array can sequentially acquire an image based on a clock pulse internal to the camera or data acquisition electronics. In this case, the positional encoder is used to precisely control the speed of the scanning mechanism and line image acquisition is synchronized to provide a desired number of pixels per unit length in the direction of scanning, thus achieving high-resolution imaging and the stability needed for precision measurements.
More particularly, in order to compensate for repeatable non-linearities and inconsistencies in scanner motion, a high precision optical reticle comprised of a number of fiducial indices is employed. The reticle is optically projected onto a portion of the image plane where the linear array is scanning. Mor
Handal & Morofsky
Mehta Bhavesh
LandOfFree
High precision three dimensional mapping camera does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with High precision three dimensional mapping camera, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and High precision three dimensional mapping camera will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2860112