Accurate RGB-D camera based on structured light techniques

Recently, RGB-D cameras have been widely used to perform 6-D pose estimation of target objects for robotic manipulation. Such applications require accuracy shape measurements for 3-D modeling of the objects. In this work, we develop an RGB-D camera based on the structured light (SL) techniques with gray-code encoding and decoding. The intrinsic and extrinsic parameters of the camera system are determined by a calibration process. 3-D reconstruction of the object surface is based on a ray triangulation principle by computing the intersection point. In the implementation, we construct an RGB-D sensing system with an industrial camera and a digital light projector. In the experiments, real world objects are used to test the feasibility of the proposed method. The evaluation carried out using planar objects has demonstrated the accuracy of our RGB-D camera system.