Curve and Circle Fitting of 3D Data Acquired by RGB-D Sensor

Abstract

In robot vision systems, three-dimensional data reconstruction becomes popular to be used in new research procedures. There are different types of depth sensors in industrial applications. These sensors are easy to access the data while capturing. RGB-D sensor obtains the three-dimensional point cloud data from one single shot. Xtion 2 depth sensor detects all point cloud data clearly which can be seen by sensor. Therefore, it is very important to observe the high precision results while using in grasping applications. We developed the curve and circle fitting of 3D data acquired by the sensor. In this system, the 3D point cloud data is examined by curve and circle estimation from sensor. Moreover, the adjusted R-square Regression Model is applied to verify the accuracy results between estimated data and actual data. This system is implemented by using Visual C++ language in Microsoft Visual Studio and using Microsoft Foundation Class (MFC) window form application. In implementation of measurement system, the pig model is used in measurement area.