Apparatus and method for processing object data in computed tomography data using object projections

• US 6,345,113 B1
• Filed: 01/12/1999
• Issued: 02/05/2002
• Est. Priority Date: 01/12/1999
• Status: Expired due to Term

• Current Assignee: Analogic Corporation
• Original Assignee: Analogic Corporation
• Inventors: Simanovsky, Sergey, Hiraoglu, Muzzafer, Crawford, Carl R., Bechwati, Ibrahim M.
• Primary Examiner(s): Au, Amelia M.
• Assistant Examiner(s): MILLER, MARTIN E
• Application Number:
  US09/228,379
• Time in Patent Office: 1,120 Days
• Field of Search: 382/131, 382/206, 378/4, 378/5, 378/15, 378/57, 378/21, 378/901, 378/62, 250/363.04
• US Class Current: 382/131
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• CPC Class Codes: G06V 2201/05 Recognition of patterns representing particular kinds of hidden objects, e.g. weapons, explosives, drugs

Abstract

An apparatus and method for identifying and classifying objects represented in CT data for a region using the shape of the objects are disclosed. An object represented by CT data for a region is identified. Next, a two-dimensional projection of the object along a principal axis of the object is generated. The principal axis can be identified by computing eigenvectors of a covariance matrix of spatial locations of voxels in the CT data that are associated with the object. The smallest eigenvector can be selected as the principal axis of the object along which the two-dimensional projection is generated. The identification of the object can be used in the classification of the object such as by altering one or more discrimination parameters.

Claims

1. A method of detecting objects represented in three-dimensional reconstructed computed tomography (CT) data acquired within a three-dimensional region defined by a three-dimensional scanning coordinate system, comprising:

identifying object volume elements from the volume elements of the three-dimensional reconstructed CT data so as to define volume elements constituting an entire object;

identifying three axes defined by eigenvectors associated with the entire object; and

generating a two-dimensional projection from three-dimensional reconstructed CT data of the entire object in a plane associated with one of the three axes defined by the eigenvectors associated with the entire object;

wherein the plane in which the two-dimensional projection is generated is a plane in which the two-dimensional projection has maximum area;

wherein the plane in which the two-dimensional projection has maximum area is identified by searching solid angles over a hemisphere in the three-dimensional region defined by the three-dimensional scanning coordinate system.

2. A method of detecting objects represented in three-dimensional reconstructed computed tomography (CT) data acquired within a three-dimensional region defined by a three-dimensional scanning coordinate system, comprising:

identifying object volume elements from the volume elements of the three-dimensional reconstructed CT data so as to define volume elements constituting an entire object;

identifying three axes defined by eigenvectors associated with the entire object; and

generating a two-dimensional projection from three-dimensional reconstructed CT data of the entire object in a plane associated with one of the three axes defined by the eigenvectors associated with the entire object;

wherein the two-dimensional projection of the object comprises a plurality of projection pixels;

each projection pixel is associated with a density value; and

the density value of each projection pixel corresponds to a count of a number of object volume elements in a direction of the axis defined by the object, each object volume element in a count being weighted by a density value associated with the volume element.

3. An apparatus for detecting objects represented in three-dimensional reconstructed computed tomography (CT) data acquired within a three-dimensional region defined by a three-dimensional scanning coordinate system, comprising:

an object identifier that (a) identifies object volume elements from the volume elements of the three-dimensional reconstructed CT data so as to define volume elements constituting the entire object;

(b) identifies three axes defined by eigenvectors associated with the entire object; and

(c) generates a two-dimensional projection from three-dimensional reconstructed CT data of the entire object in a plane associated with one of the three axes defined by the eigenvectors associated with the entire object;

wherein the plane in which the two-dimensional projection is generated is a plane in which the two-dimensional projection has maximum area; and

the plane in which the two-dimensional projection has maximum area is identified by searching solid angles over a hemisphere in the region defined by the three-dimensional scanning coordinate system.

4. An apparatus for detecting objects represented in three-dimensional reconstructed computed tomography (CT) data acquired within a three-dimensional region defined by a three-dimensional scanning coordinate system, comprising:

an object identifier that (a) identifies object volume elements from the volume elements of the three-dimensional reconstructed CT data so as to define volume elements constituting the entire object;

(b) identifies three axes defined by eigenvectors associated with the entire object; and

(c) generates a two-dimensional projection from three-dimensional reconstructed CT data of the entire object in a plane associated with one of the three axes defined by the eigenvectors associated with the entire object;

wherein the two-dimensional projection of the object comprises a plurality of projection pixels;

each projection pixel is associated with a density value; and

the density value of each projection pixel corresponds to a count of a number of object volume elements in a direction of the axis defined by the object, each object volume element in a count being weighted by a density value associated with the volume element.

5. An apparatus for detecting objects represented in three-dimensional reconstructed computed tomography (CT) data acquired within a three-dimensional region defined by a three-dimensional scanning coordinate system, comprising:

means for identifying object volume elements from the volume elements of the three-dimensional reconstructed CT data so as to define volume elements constituting an entire object;

means for identifying three axes defined by eigenvectors associated with the entire object; and

means for generating a two-dimensional projection from three-dimensional reconstructed CT data of the entire object in a plane associated with one of the three axes defined by the eigenvectors associated with the entire object;

wherein the plane in which the two-dimensional projection is generated is a plane in which the two-dimensional projection has maximum area; and

the plane in which the two-dimensional projection has maximum area is identified by searching solid angles over a hemisphere in the three-dimensional region defined by the three-dimensional scanning coordinate system.

6. An apparatus for detecting objects represented in three-dimensional reconstructed computed tomography (CT) data acquired within a three-dimensional region defined by a three-dimensional scanning coordinate system, comprising:

means for identifying object volume elements from the volume elements of the three-dimensional reconstructed CT data so as to define volume elements constituting an entire object;

means for identifying three axes defined by eigenvectors associated with the entire object; and

means for generating a two-dimensional projection from three-dimensional reconstructed CT data of the entire object in a plane associated with one of the three axes defined by the eigenvectors associated with the entire object;

wherein the two-dimensional projection of the object comprises a plurality of projection pixels;

each projection pixel is associated with a density value; and

the density value of each projection pixel corresponds to a count of a number of object volume elements in a direction of the axis defined by the object, each object volume element in a count being weighted by a density value associated with the volume element.