This paper focuses on the design of an inexpensive and accurate range scanner for automatic acquisition of a CAD model of a manufactured part by using two‐dimensional…
This paper focuses on the design of an inexpensive and accurate range scanner for automatic acquisition of a CAD model of a manufactured part by using two‐dimensional images to determine a digitized three‐dimensional shape. In the developed approach, the object is passed at a speed of 4 cm/s through a single linear laser stripe and forty continuous images are captured into the frame memory of the host computer for subsequent processing. A major problem that is encountered in the design of laser stripe scanner is the specula reflection, which can be mitigated by the developed approach. Six center‐locating algorithms are described, which are central to the developed approach. These algorithms are able to achieve sub‐pixel accuracy. The center of mass algorithm that uses three points, gives the best repeatability over the other algorithms. The center of mass algorithm that uses intensity threshold, provides the best linearity over the other algorithms.
This paper aims to address three major issues in the development of a vision‐based navigation system for small unmanned aerial vehicles (UAVs) which can be characterized…
This paper aims to address three major issues in the development of a vision‐based navigation system for small unmanned aerial vehicles (UAVs) which can be characterized as follows: technical constraints, robust image feature matching and an efficient and precise method for visual navigation.
The authors present and evaluate methods for their solution such as wireless networked control, highly distinctive feature descriptors (HDF) and a visual odometry system.
Proposed feature descriptors achieve significant improvements in computation time by detaching the explicit scale invariance of the widely used scale invariant feature transform. The feasibility of wireless networked real‐time control for vision‐based navigation is evaluated in terms of latency and data throughput. The visual odometry system uses a single camera to reconstruct the camera path and the structure of the environment, and achieved and error of 1.65 percent w.r.t total path length on a circular trajectory of 9.43 m.
The originality/value lies in the contribution of the presented work to the solution of visual odometry for small unmanned aerial vehicles.