Computer graphics processing and selective visual display system – Computer graphics processing – Three-dimension
Reexamination Certificate
1997-11-13
2002-06-18
Razavi, Michael (Department: 2672)
Computer graphics processing and selective visual display system
Computer graphics processing
Three-dimension
C345S421000, C345S581000
Reexamination Certificate
active
06407735
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to a method for analyzing and ordering informational data representing the physical characteristics of an object. More specifically, the invention relates to the generation of ordered point cloud information from layered input data and the output of a surface-based object representation.
Various apparatus and methods exist in the art for reverse engineering an existing product in order to determine its structure and makeup. Such a reverse engineering process may be desirable because a computer aided design file representing the object does not exist, because a mold for the object does not exist, or simply because the part has been modified independently of the existing rebuilt materials.
It is often desirable to know the geometry, both external and internal, of existing objects. Various apparatus for the delamination of such objects are known in the art. For example, such an apparatus and method for creating three-dimensional modeling data from an object is disclosed in U.S. Pat. No. 5,621,648, to Crump, which is hereby incorporated by reference in its entirety. The '648 patent discusses and discloses the delamination of an object encased in an encasing material by successively cutting off layers of the encased object, imaging the exposed surfaces and producing a surface model by lofting surfaces on a stack of generated spline art layers. Such a method and apparatus allows the object to be examined for both external and internal geometry.
Other methods and apparatus for traditional measuring of dimensions and generation of surfaces involve extensive use of a probe and metering technique by which a physical probe is placed at numerous and various points on an object, making measurements of the relative location at each touching point. Such a process is tedious and is subject to the human error of misplacing the probe or failing to recognize important changes in geometry of a system. Even if such a probe and meter system were automated, the time consumed by such a system would be immense due to the large numbers of points to be probed. Further, such a probe and metering technique does not capture internal geometry well, especially internal geometry with no external availability. Such geometry will not be captured by a probe and meter technique.
Delaminating systems and other systems which generate or provide physical characteristic data of various objects, such as magnetic resonance imaging (MRI), X-ray technology and the like, use a wide variety of output file formats and output protocols to transmit raw data information pertaining to the object to a computer aided design tool or other computer system for construction of a three-dimensional image representative of the object. Many such systems export a continuous stream of data, and many surfacing computer programs, commonly used to work with such data sets require input of all data at one time. The vast amount of data generated by such a process requires a great deal of storage space and computer power in order to be able to work with it. Too much data leads to an unwieldy process. Working with a full data set of information creates numerous problems with the usefulness of such data.
The delaminating machine disclosed in the '648 patent and other object representation technologies can provide data generated layer by layer. When working with layered data, smaller amounts of information are used at one specific time, reducing the time and storage requirements for image and object reconstruction. It would be desirable to be able to use layer by layer data from a delamination or other imaging system to construct a computer aided design image or specific output file format which is more usable because it has a reduced size without sacrificing informational content.
A further problem with other systems or methods for generating three-dimensional output data lies in their inability to appropriately and accurately distinguish between internal and external geometry configurations. Capturing a complete data set incorporating all the geometry of an object has several requirements, including the gathering of enough data points to define a surface, and measurement and sensing capable of providing data representative of all surfaces. If the surface to be analyzed or reconstructed is obstructed, such as an internal surface or the inside of a specific part, important geometry will be difficult or impossible to measure or photograph, and will be especially difficult to distinguish using probe and meter techniques. Further, a method is required to distinguish between external and internal geometries to be able to accurately generate surface representations. A system such as that disclosed in the '648 patent provides data representative of all internal and external geometry. However, once complete internal and external geometry representations have been captured, or generated, the amount of data required to accurately represent such surfaces can be extreme given the resolution of the generated representations. It would be desirable to provide a method which could accurately and appropriately distinguish between internal and external geometry points.
When layer by layer data is sent in such a layered order to a software program or integrated hardware and software system for generation of an output file, it may take many formats. For example, in the '648 patent, each scanned surface layer will typically contain a series of pixels of either black or white, with one color representing encasing material, and the other pixel color representing the object. Other imaging technologies may also provide layered data representative of object and non-object pixels. It would be desirable to provide a method of taking data representations of this type and for processing them to produce a three-dimensional surface representation while reducing the amount of information without reducing the quality of the final generated image.
Layer by layer data will contain certain object geometries which may be difficult to recognize. Such geometries include the first or last layer, referred to as a cap, on which a geometry appears. This may be an external geometry or an internal geometry. Also, at certain levels or layers throughout an object, especially a complex object, geometries may split. For example, on one layer there may be one section of a whole, and on the next, there may be two sections of the same whole that have split from the first section of the whole. A table fork for example splits from the handle to a series of tines. It has been difficult for prior art methods to distinguish a split, or conversely, a join, in layer by layer geometry. It would, therefore, be desirable to provide a method by which caps, splits, and joins are recognized and accurately dealt with. It would also be desirable to provide a method which would convert layer by layer pixel data to point data which may be more suitably used by a data processor such as a computer and associated software.
SUMMARY OF THE INVENTION
It is an object of the present invention to provide a method for generating ordered point cloud information based on layer by layer data.
It is another object of the present invention to improve a method for recognition and treatment of caps, splits, and joins in three dimensional object representation data.
It is still another object of the present invention to provide a method for selective sampling of points and layers in a layer by layer representation of a three-dimensional object.
It is yet another object of the present invention to provide a method for generation of three-dimensional surface representations of an object by stitching layered point clouds together to form a series of facets.
The present invention achieves the preceding objects and overcomes the problems of the prior art by providing a method of analyzing, refining and working with layer by layer pixel data information generated by a delaminating machine or the like to provide an output format representative of a three dimensio
Crump Group Inc.
Good-Johnson Motilewa
Moore & Hansen
Razavi Michael
LandOfFree
Method for generating surface representations of objects... 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 for generating surface representations of objects..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for generating surface representations of objects... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2969692