A new method for registration of heterogeneous sensors in a dimensional measurement system

Abstract

Registration of multiple sensors is a basic step in multi-sensor dimensional or coordinate measuring systems before any measurement. In most cases, a common standard is used to be measured by all sensors, and this may work well for general registration of multiple homogeneous sensors. However, when inhomogeneous sensors detect a common standard, it is usually very difficult to obtain the same information, because of the different working principles of the sensors. In this paper, a new method called multiple steps registration is proposed to register two sensors: a video camera sensor (VCS) and a tactile probe sensor (TPS). In this method, the two sensors measure two separated standards: a chrome circle on a reticle and a reference sphere with a constant distance between them, fixed on a steel plate. The VCS captures only the circle and the TPS touches only the sphere. Both simulations and real experiments demonstrate that the proposed method is robust and accurate in the registration of multiple inhomogeneous sensors in a dimensional measurement system.

A Globe Calibration Method for Optical Multisensor in 3D Complex Surface Measurement System

“…At present, there are two types of methods for calibration of the global coordinate system. One adopts a unified calibration object according to the different types of sensors and realizes the global calibration by measuring the calibration object with each sensor [8][9][10]. The other adopts the numerical calculation of the rigid-body transformation matrix by extracting three-dimensional data characteristics of the same object measured by each sensor to achieve global calibration [11,12].…”

“…A modified iterative closest-point algorithm has been proposed to improve the registration performance for multisensory coordinate metrology using curvature information [21]. Zhao et al [8] adopted separated standards to be measured by a video camera sensor and a tactile probe sensor. A common reference coordinate system was created adopting a multistep registration method to match the datasets of two inhomogeneous sensors.…”

Mapping Based Calibration for Integrated Measuring System on Reverse Engineering of Freeform Shape

“…Before the measuring step, calibration is always important. Since each component has its own coordinate system, transformation must be undertaken to align the components into a coherent coordinate system [19,25]. There are many proposed approaches on calibration computing [12][13][14][15]17], such as linear or non-linear optimization techniques, two-step technique.…”

Design of Insertion Mechanism for Cannula Flexible Needle Based on TRIZ

“…Pang et al [20] proposed the two-step optimization method to calibrate the turntable axis with a 3D scanner, an iterative procedure was used to optimize the distance in the registered pairs. Instead of analyzing the measured 3D data, some approaches are based on digitizing calibration targets of known dimensions [22,23,25] such as tooling balls. While Zhang [22] used a cylindrical standard to calibrate the Three-Microscope Vision System.…”