3-D shape measurement of complex objects by combining color-coded fringe and neural networks

A new approach for three dimensional (3-D) shape measurement was proposed based on colorcoded fringe and neural networks. By applying the phase-shift technique to fringe projection, point clouds were generated with high spatial resolution and limited accuracy. The picture element correspondence problem was solved by using projected color-coded fringes with different orientations. Once the high accurate corresponding points were decided, high precision dense 3-D points cloud was calculated by the well trained net. High spatial resolution can be obtained by the phase-shift technique and high accuracy 3-D object point coordinates are achieved by the well trained net, which is not dependent on the camera model and will work for any type of camera. Some experiments verify the performance of this method.