New 3-D Imaging Capabilities Announced By FIRSTSIGHT VISION

The C3 family of robust laser scanning cameras from Automation Technology GmbH offers outstanding speed and resolution for 3-D imaging applications. On-board FPGA pre-processing of the raw image data enables the C3 3-D cameras to provide previously unattainable frame rates for inspection, measurement and sorting in the third dimension.

C3 cameras are designed for use with laser line illumination for 3-D imaging and have a resolution of 2352 x 1728 pixels. Extremely fast profile measurement speeds of up to 25 KHz deliver up to 58 million 3-D points per second. By evaluating the laser line data within the camera, only the information that is relevant to the application is determined, such as profile height and intensity values, which can then be sent over the interface to a further processing system such as the Common Vision Blox, imaging toolkit from STEMMER IMAGING. This integrated processing determines the height profile of the measured surface with a sub-pixel precision of 1/16 pixel without losing speed.

Common Vision Blox now features algorithms for capturing and comparing 3-D objects and their surfaces developed by the Spanish company, AQSENSE. The AQSENSE Shape Processor software allows the alignment and comparison of dense 3D clouds of points with their respective models. Based on best-fit approach, the patent-pending alignment procedure relies on a very fast and accurate technology, which combined with an optimized comparison engine, operates on huge clouds of points in a fraction of a second. In practical terms, it compares the 3D-image data of a good part - the so called golden template - with the upcoming test parts. One of the most impressive features is that the tool aligns the test parts image with the template image in 3D before the comparison. Although this is a major calculation, the accuracy and processing speed of the algorithms mean that the tool can be used for real in-line production processes.

The C3 family is available in CameraLink and Gigabit Ethernet versions for straightforward integration into both new and existing systems. In addition, the CameraLink models uniquely include a GeniCam Transport Layer (normally found only on GigE Vision cameras) for DALSA X64, Xcelera, PCCamlink, PC2Camlink, Euresys Grablink, Matrox Meteor and National Instruments framegrabbers. This means that cameras on these framegrabbers can be *detected*, *configured* and *accessed* in exactly the same way as a GigE Vision camera.

For added versatility, a multi-Region Of Interest readout on the sensor allows output of an area-scan image, a 3-D image and a scatter image, allowing for 2-D and 3-D image comparison.