Three-dimensional surface reconstruction using optical flow for medical imaging

Abstract

The recovery of a three-dimensional (3-D) model from a sequence of two-dimensional (2-D) images is very useful in medical image analysis. Image sequences obtained from the relative motion between the object and the camera or the scanner contain more 3-D information than a single image. Methods to visualize the computed tomograms can be divided into two approaches: the surface rendering approach and the volume rendering approach. In this paper, a new surface rendering method using optical flow is proposed. Optical flow is the apparent motion in the image plane produced by the projection of the real 3-D motion onto 2-D image. The 3-D motion of an object can be recovered from the optical-flow field using additional constraints. By extracting the surface information from 3-D motion, it is possible to get an accurate 3-D model of the object. Both synthetic and real image sequences have been used to illustrate the feasibility of the proposed method. The experimental results suggest that the proposed method is suitable for the reconstruction of 3-D models from ultrasound medical images as well as other computed tomograms.

Advancing computer-assisted orthopaedic surgery using a hexapod device for closed diaphyseal fracture reduction

“…During image processing, we introduced a threedimensional (3D) surface stereolithography (STL) model of the fractures as our reconstruction method, based on computed tomography (CT) data from Mimics software v. 10.01 (Materialise NV, Leuven, Belgium), using the surface rendering algorithm of marching cubes (16,17) to accurately measure and analyse the geometric parameters of the bones. The 3D image was reconstructed from DICOM slices of high-resolution CT scans made of point cloud data, whose surface consists of characteristics of the femur anatomy.…”

Design of Person Flow Counting and Monitoring System Based on Feature Point Extraction of Optical Flow

“…Corner is usually considered the point of intersection of two edges, and it can also be regarded as feature points with the same direction in a small neighborhood. Moreover, these points can obtain high gradient in multiple directions [5] . The principle of extracting corner of the image is that generally exists in the image gradient in multiple directions simultaneously, on the basis of Moravec operator, Harris focus detection method was proposed in 1998 by Chris Harris and Mike Stephens.…”

Robust point correspondence matching and similarity measuring for 3D models by relative angle-context distributions

“…The ability to establish robust correspondence between 2D images and more recently, 3D objects is frequently a central issue in many image analysis paradigms, such as structure-from-motion [1], stereoposis [2], optical flow analysis [3], and deformable or morphable models [4,5], etc. This is particularly true in the medical and biological domains where robust solutions to the correspondence problem is vital to many applications such as organ growth measurement [4], histological section alignment [6], aesthetic surgical planning [7].…”