Block-detection and three-dimensional moment algorithms are applied to determine the presence and orientation of cylinders in a three-dimensional image. The proposed method detects blood vessels in a three-dimensional tomographic image constructed by infrared radiometry. Based on the moment principle, the method can be employed to determine the centroid, diameter, and orientation of arbitrarily shaped blood vessels for application in the laser treatment of hypervascular skin lesions. © 1998 Elsevier Science Ltd.

A three-dimensional Tomographic Reconstruction Algorithm (3D TRA) has been developed and implemented on a parallel machine at the Maui High Performance Computing Center (MHPCC). Given a time sequence of infrared images, this algorithm provides information to determine the location, physical dimensions, and initial space-dependent temperature increase of laser heated blood vessels. In this approach the serial computation of a 3D TRA is decomposed into multiple tasks which can be executed in parallel to reduce the computation time.