Three-Dimensional Measurement Method of Four-View Stereo Vision Based on Gaussian Process Regression

Abstract

Multisensor systems can overcome the limitation of measurement range of single-sensor systems, but often require complex calibration and data fusion. In this study, a three-dimensional (3D) measurement method of four-view stereo vision based on Gaussian process (GP) regression is proposed. Two sets of point cloud data of the measured object are obtained by gray-code phase-shifting technique. On the basis of the characteristics of the measured object, specific composite kernel functions are designed to obtain the initial GP model. In view of the difference of noise in each group of point cloud data, the weight idea is introduced to optimize the GP model, which is the data fusion based on Bayesian inference method for point cloud data. The proposed method does not require strict hardware constraints. Simulations for the curve and the high-order surface and experiments of complex 3D objects have been designed to compare the reconstructing accuracy of the proposed method and the traditional methods. The results show that the proposed method is superior to the traditional methods in measurement accuracy and reconstruction effect.

Smart optical coordinate and surface metrology

“…Further to the publication of these review works, there have been many studies conducted on development of machine learning methods for data fusion in optical metrology, used for applications ranging from micro-scale surface texture measurement to mapping (for example [104][105][106][107]). One trend in research in the field of multi-sensor data fusion is improvement of the functionality of traditional algorithms with machine learning or other artificial intelligence (AI) techniques.…”

“…One trend in research in the field of multi-sensor data fusion is improvement of the functionality of traditional algorithms with machine learning or other artificial intelligence (AI) techniques. For example, Gong et al [105] proposed an machine learning algorithm based on Gaussian process regression, using a fourview stereo vision system to gather data. Gong et al tested their new method using a high-order curved surface and experiments with a freeform surface, with results indicating that this Gaussian process-based machine learning method performed better than traditional Gaussian process-based algorithms, in terms of the accuracy and efficiency of the reconstruction.…”

Application of Computer Vision in T-Shirt Dimensions Measurement

“…In [5][6][7][8] the authors reported new studies to be able to test most of the defects that might appear during actual production. To aid in the checking of dimensions, stereo vision-based measurement methods have been developed and applied in many fields, including distance measurement, deformation measurement, and object reconstruction of 3-dimensional objects [9][10][11][12]. Computer vision systems that combine multiple cameras or combine a camera with multiple sensors are also widely used in measuring the size of objects in industrial applications [13], [14].…”

A Multisensor Data Fusion Method Based on Gaussian Process Model for Precision Measurement of Complex Surfaces

“…However, such methods only substitute the data from different sources locally and cannot improve the accuracy of the overall data. Therefore, the data fusion methods based on modern statistical theory have received much attention [20][21][22][23]. The core idea of such methods is to use statistical methods to model the measurement datasets obtained from different sensors, and then provide the final prediction of each position.…”