Arbitrary viewpoint three-dimensional imaging method using compressed voxel data constructed by a directed search of voxel data representing an image of an object and an arbitrary viewpoint

• US 5,644,689 A
• Filed: 01/06/1993
• Issued: 07/01/1997
• Est. Priority Date: 01/13/1992
• Status: Expired due to Term

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Suzuki, Ryuichi, Ban, Hideyuki
• Primary Examiner(s): Herndon, Heather R.
• Assistant Examiner(s): Fetting, Anton W.
• Application Number:
  US08/001,389
• Time in Patent Office: 1,637 Days
• Field of Search: 395/119, 395/120, 395/122, 395/123, 395/124, 395/126, 395/127, 395/136, 395/137, 395/138
• US Class Current: 345/424
• CPC Class Codes: G06T 15/10 Geometric effects

Abstract

Original image coordinate data representing boundary positions of an object detected through the search of the object along predetermined multiple directions is produced in advance. A range image seen from a display plane is produced through the affine transformation process for the detected boundary positions and comparison and substitution processes for distances by means of a Z-buffer. A shading process is performed based on the range image, and the resulting object image is displayed on a CRT screen or the like. This method enables the transformation of voxel-wise original image data into original image coordinate data merely through the sequential access to the original image data, eliminating the need of intricate image processes such as tracking on the object surface, and is capable of displaying a three-dimensional image of the object quickly and also enhancing the picture quality of the displayed image.

Claims

1. An arbitrary viewpoint three-dimensional imaging method using compressed voxel data constructed by a directed search of voxel data representing an image of an object, said method comprising the steps of:

(a) searching at least six neighboring voxel data positions around each of all voxel data inside of said object represented by the voxel data along each of six axes directions in a space defined by three coordinate axes intersecting at right angles with each other to detect boundary voxel positions representing surface positions of said object in each of the six axes directions, the image of said object being represented in the three coordinate axes;

(b) producing eight kinds of compressed boundary coordinate data representing boundary voxel positions from eight viewpoint positions in the three coordinate axes having different voxel connectivity from the neighboring voxels;

(c) storing, in a memory, the compressed boundary coordinate data from the eight viewpoint positions for each difference of voxel connectivity in the three coordinate axes;

(d) selecting one kind of the compressed boundary coordinate data from the memory based on positive or negative signs of three coordinate axes components of a vector which is normal to a display plane and points from the display plane to the object;

(e) transforming the boundary voxel positions expressed by the selected compressed boundary coordinate data into coordinate values on the display plane and distances therefrom to the display plane;

(f) producing, from the transformed boundary voxel positions, range data representing the distances from the display plane to the boundary voxel positions based on the viewpoint position using a Z-buffer;

(g) shading a surface of said object based on said range data; and

(h) displaying a three-dimensional image of said object based on the shaded surface.