A fast algorithm for decoding an object image in structured light for measuring a three-dimensional profi le in a non-linear optical path

Abstract

The work is devoted to the development of image processing algorithms for measuring a three-dimensional surface profi le using phase triangulation methods. An algorithm for decoding images of an object in structured light is proposed, based on an iterative search for the minimum deviation of the model function from the measurement results and compensating for the nonlinearity of the receiving-transmitting path of the measuring complex. A distinctive feature of the proposed method is that the search for the model function is sought in the form of a polynomial of the second degree, which ensures the stability of the method against nonlinear distortions caused by power-law transformations of the receiving-transmitting path. The use of the interval search method as part of the proposed algorithm provides a qualitative reduction in its algorithmic complexity. It is shown that the proposed algorithm provides a stable search for the value of the initial phase shift in object images, is resistant to noise and non-linear distortions. The algorithm proposed in the paper can be used to process images received in three-dimensional scanning systems based on triangulation methods using structured illumination and phase triangulation. The proposed algorithm will be especially useful for data processing during the operation of measuring systems in the conditions of a nonlinear source-receiver path of optical radiation and measurement time limitations.