One of the hot emerging technologies in the automotive industry is the collision avoidance system which detects an imminent frontal crash using radar, optical sensors, or a combination thereof, and engages warning, braking and/or steering systems in response. Similarly, autonomous robot manufacturers are using 3D object detection and tracking to guide the motions of robots or robotic vehicles. The security and surveillance industries also make use of this technology to detect and track motion in a video sequence. With the release of HALCON 12, MVTec now provides machine vision system developers the ability to detect and track the 3D motion of objects with HALCON’s new 3D scene flow.
3D scene flow works using two consecutive stereo image pairs to determine uncalibrated (relative) and calibrated (real-world coordinate) 3D motion data. In the uncalibrated mode, 3D scene flow provides relative motion information in the form of optical flow (the 2D shift of features between consecutive images) plus the change in disparity of features between two consecutive stereo image pairs. Calibrated 3D scene flow provides 3D motion data by detecting and tracking the 3D position changes of surface points on a 3D object model found in consecutive point cloud scenes. In this case, the point clouds are reconstructed from the stereo image pairs.
These example images show how 3D scene flow might be applied to determine the 3D motion of objects (automobiles) in a dynamic environment and the rotational axis and the angle of rotation of a rigid 3D object.
You can view an example program demonstrating 3D scene flow by downloading the HALCON 12 software for free and viewing the scene_flow_globe.hdev example.
Multipix Imaging are the UK distributor for HALCON 12 with trained support staff, please call us for further details on 01730 233332.