Based On Structured Light Complex Object Profile 3D Measurement

This paper study and discuss complex object three-dimensional vision measurement profilometry system based on the structured light technique which consists of micro-computer,digital projector and video camera. Structured light measurement system is similar to multi-camera stereoscopic system. And the difference between of them is that one of camera is instead of projector. This technique greatly simplifies the problem of the correspondence three-dimensional vision measurement, that the correspondence between the same point in different images.Now, structured light technique has become a hot topic in the area three-dimension vision measurement. There are many structured light methods proposed by a lot of domestic and foreign scholars, which have been a reliable and stable three-dimensional measurement system. Depending on the coding strategy, the structured light can be divided into three categories: time coding, spatial neighborhood coding and direct coding. In this paper, wo study high precision structured light 3D vision measurement technique based on sinusoidal fringe grating encoding method in time coding strategy.For the actual measurement process,projection sinusoidal gratings and captured images are disturbed by various, which lead to the image not only including useful fringe information but also including electronic noise or other interferences then reducing the precision of measurement system. Thus, this paper focuses on the method of nonlinear gamma distortion correction algorithm and clearing the noise interference method in captured images so that implement high-precision measurement of object profilometry. The main contents, in this paper, are as follows:Firstly, we analysis the fringe projection profilometry the main structured and principle,then establish the phase-depth mathematical models and the method of extracting wrapped phase. According to build mathematical models, find noise source or the reason causing interferences in the measurement, and analysis them.Secondly, through theoretical analysis, we find the main influence precision factors are high-frequency noise of the system and nonlinear gamma distortion. This paper utilizes the spin filter instead of traditional low pass filter to clear high-frequency noise in captured fringe images, which can clearly eliminate high-frequency noise of system at the same time advise harm the fringe feature in captured images.Thirdly, build nonlinear gamma distortion mathematical model and analysis detail influence of gamma distortion on precision of the fringe projection measurement system. For nonlinear gamma distortion characteristic,we study two methods to correct gamma distortion so that improve system measurement accuracy. First method is based on iteration phase compensation correction algorithm to correction gamma error, which has advantage of easy method and correction fast. Another method is uniform fringe pattern to correct phase error.However, as projecting many uniform fringe patterns in correction stage, this method is more complex than the first one.Finally, we through many experiments to verify the effectiveness of the algorithm proposed in measurement accuracy improving. This paper uses four step phase-shifting methods,and phase-shifting value of four fringe patterns are 0°,900°?180°?270°. The measurement objects are flat and sculpture. Then, we uses correction method in this paper to correct experimental data and get the experimental results. Finally, according to experimental results, we discuss the advantages and disadvantages of proposed gamma correction algorithm.