Frame Grabber Power for Photometric Stereo Method
It is nearly impossible to distinguish an object’s texture from the surface properties in a single image. With the photometric stereo approach, at least three and usually four images are taken with a camera where the object is illuminated from different directions. Four images (texture, X- and Y-derivation and average curvature) are typically calculated from these captures. This separation of the surface properties from the texture enables a clean evaluation of the products.
Due to the time-consuming calculation in the CPU the required cycle times in production cannot usually be met. MSTVision GmbH, a development service provider for tailor-made solutions in the field of industrial image processing, has used a frame grabber in its photometric stereo approach and was thus able to significantly accelerate the implementation of image acquisition and calculation in the process.