Mathematical models for metric features extraction from RGB-D sensor

The use of the RGB-D camera has been applied in several fields science. That popularization is due to emergence technologies such as Intel® RealSense™ D400 series. However, despite actual demand from some potential users, few studies concern characterization these sensors for object measurements. Our study sought estimate models dealing with calculating area and length between targets or points within RGB depth images. An experiment was set up white cardboard fixed on a flat surface colored pins. We measured distance by average pixels belonging target area. Information Criterion AIC BIC associated R2 were performed select best models. Polynomial power regression reached highest coefficient determination smallest values BIC.