Boots – shoes – and leggings
Patent
1989-10-17
1993-09-14
Envall, Jr., Roy N.
Boots, shoes, and leggings
36441319, 36441315, G06F 1542, A61B 603
Patent
active
052455386
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
An object of the present invention is a process for representing views of an object. This object has to be discriminated in a digital volume. A digital volume results from the acquisition and storage of pieces of information concerning physical characteristics of internal parts of an examined body. The best known mode of acquisition of this information is the tomographic mode. This mode may be performed, especially in the medical field, by various means: nuclear magnetic resonance, tomodensitometry by X-rays, or tomography by gammagraphy or ultrasonic tomography. This acquisition mode may, of course, also be any mode, once it leads to a gathering of pieces of physical information that can be arranged virtually with respect to one another along three orthogonal axes of reference and are supposed to represent, in the virtual position that they occupy, the physical characteristics of corresponding parts in an examined body. Each volume element of this digital volume thus has at least two types of information. A first type concerns the coordinates of a corresponding place in the body (corresponding biuniquely to an address in a storage memory of a computer). The other type represents the value of the physical information which has been assigned to this place and which has been measured, for example by one of the above-mentioned techniques of tomography.
Techniques for the representation of views of objects to be discriminated in digital volumes are already known. There are chiefly two of them. A first technique, developed mainly by G. T. HERMAN and his team, comprises the sequence of the following operations. Taking a given digital volume, reformatted if necessary so that the resolution is identical in all three spatial dimensions, first of all a segmentation is done. The principle of the segmentation consists in comparing the values of the physical information loaded in each of the volume elements with a reference value and in selecting those of these volume elements for which the value of the physical information belongs, for example, to a value range located around this reference value. To simplify the matter, in tomodensitometry, it can be understood that a test on the density will enable differentiation, in the digital volume, of the volume elements corresponding to bones (high density) and the volume elements corresponding to soft tissue (low density). It is then possible to have a collection of addresses of memory cells that correspond to chosen volume elements and the set of which defines the object thus segmented.
The principle of the representation then consists in attributing a visible surface to each of the chosen volume elements (which, however, are located on the surface of the segmented object), computing an orientation of this surface (in estimating an orientation of the normal to it), and assessing the luminous contribution of this surface to an image of a view when this surface is exposed to a given illumination (namely, to an illumination coming, for example, from a precise point of the space external to the segmented object). The locations of the chosen volume elements, and hence of the corresponding visible surfaces being known, they can be attributed, in the image of the view to be represented, elements of the surface of this image for which the coordinates, in the image, depend on the point of view from which the segmented object is looked at. These surface elements are then assigned a luminosity representing the contributions of the visible surfaces to which they are assigned. The set of all the surface elements of the image constitutes the image of the view of the object.
This process has drawbacks; in particular, the shading is not satisfactory therein. For, the image obtained shows an effect of line artefacts or circle artefacts that seem to match the contours of the segmented object, in doing so irrespectively of the orientation of the illumination, and irrespectively of the viewpoint of observation of this object. These artefacts of contours are particu
REFERENCES:
patent: 4729098 (1988-03-01), Cline et al.
patent: 4737921 (1988-04-01), Goldwasser et al.
patent: 4751643 (1988-06-01), Lorensen et al.
patent: 4821210 (1989-04-01), Rumbaugh
patent: 4866612 (1989-09-01), Takagi
patent: 4907152 (1990-05-01), Lempriere
patent: 5095521 (1992-03-01), Trousset et al.
IEEE Transactions on Medical Imaging, vol. MI-5, No. 1, Mar. 1986, IEEE (New York, U.S.), K. H. Hohne et al.: "Shading 3D-images from CT using gray-level gradients", pp. 45-47.
Proceedings of the IEEE, vol. 71, No. 3, Mar. 1983, IEEE (New York, U.S.), J. K. Udupa: "Display of 3D information in discrete 3D scenes produced by computerized tomography" pp. 420-431.
Bodendorf A.
Envall Jr. Roy N.
General Electric CGR S.A.
LandOfFree
Process for representing views of 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 Process for representing views of an object, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Process for representing views of an object will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2032620