Invision Technologies, Inc.

Ge Homeland Protection, Inc.

Walter I Garms

Joseph Bendahan

James M Carver

A high speed computed tomography x-ray scanner using a plurality of x-ray generators. Each x-ray generator is scanned along a source path that is a segment of the scanner's source path such that each point in the source path is scanned by at least one tube. X-ray generators and detectors can be arranged in different scan planes depending on the available hardware so that a complete and near planar scan of a moving object can be assembled and reconstructed into an image of the object.

Computed tomographic scanner using rastered x-ray tubes

Systems and methods for non-intrusive analysis and display of internal features of wooden objects are provided. In embodiments of the system, a log is passed through a CT scanner in one continuous motion. One or more x-ray sources revolve around the log generating x-ray beams that traverse contiguous cross-sections of the log. An array of x-ray detectors detects x-rays that traverse the log for variations in the attenuation of rays. The detected attenuation is converted into spiral scan data that corresponds to projections in different contiguous cross-sections traversed by the x-rays. An image processor reconstructs spiral scan data into two dimensional cross-sectional images of the log by processing and formatting scan data using a planar reconstruction technique. The system renders three-dimensional views based on two-dimensional images.

Analysis and presentation of internal features of logs

A three-dimensional density distribution of a log is reduced to a two-dimensional structure that provides a convenient image or visualization of defects like knots or voids in a log, to facilitate grading and/or optimization of a sawing strategy for the log. The two-dimensional data structure is based on cylindrical or modified cylindrical coordinates Z and &thgr;. To provide a more compact identification of defects, modified cylindrical coordinates use a Z-axis that follows the growth center in the log and determines data points by evaluating properties of the log along rays at an upward angle corresponding to limbs in a tree. A process for identifying the growth center at any distance Z along the length of the log examines or accumulates the gradient of density along lines through a cross-section of the log. Manual grading and sawing optimization can employ viewing of an image based on the two-dimensional data structure with superimposed marks indicating the boundaries of faces cut from the log. Automated grading and sawing optimization employs the two-dimensional data structure to reduce processing time when compared to processes that manipulate three-dimensional data structures.

Growing community of inventors

Palo Alto, CA, United States of America

Walter I Garms