Research On Method Of 3D Vision Measurement Based On The Liner Structured Light

Machine Vision is a technology developed on the basis of the development of electronics, computer technology and signal processing theory. Vision measurement, as a very important part of Machine Vision, has been rapid developed rapidly. The meaning of vision measurement is changing the three-dimensional information of objects to the digital information which can be recognized and processed by computer directly, so as to realize the understanding of the physical characteristics of the objects. As a kind of digital conversion tool, it is convenient, rapid and accurate, and so on. Aiming at the situation of vision measurement is more and more used widely, this paper puts forward a vision measurement method assisted with liner structured light, which focus on the 3D surface reconstruction of objects, restore the objects' 3D contour, and measures the dimensions related to the objects' surface. The main research content is as follows:An image processing method which is suitable for the feature extraction of the target image is proposed in this paper. Image characteristics extraction, as a significant part in the vision measurement system, plays an important role in the accuracy of the measured results. In this paper, according to the gray-scale feature of the target images we adopt the corresponding image processing method to extract the feature information in different target images. For the grid calibration plate images, we apply an image segmentation method based on the watershed algorithm; for laser stripe images, we adopt the refinement algorithm to extract the center line of laser stripe and obtain the two feature points under certain constraints.This paper establishes a unique visual measuring system and puts forward the appropriate calibration method to calibrate the system parameters. On the basis of the traditionally visual measurement system, considering the location relations among the camera, the line structure laser and the objects, taking the light plane as one plane of the three-dimensional space coordinate system to establish the visual measurement model. After analyzing the common methods used to calibrate the camera and the light plane, applying the perspective projection transformation to calibrate the projection parameters between the plane of image coordinate system and the light plane directly, so it does not need to calibrate the internal and external parameters of the camera,as same as the parameters of the light plane.The three dimensional reconstruction and related dimension measurement of the surface of the objects is realized. In the case of the established system model and the known system parameters, obtaining the three-dimensional coordinate of the points on the surface of the objects with the help of one-dimensional sliding platform and then realizing 3D reconstruction and related dimension measurement. The experimental results show that this method is of high accuracy and small error, and it can accurately restore the 3D information of the measured objects.