Research On Calibration Technique Of Visual Measuring System

In the mechanical industry, the components are an important part of the machines. With the unceasing improvement of the quality of machine part, the method of the evaluation and the measure to the dimension of the part sets a higher request.At present, along with photo-electricity technology and computer technology unceasing development, the dimension measurement based on visual measuring technique applies more widely in the mechanical industry. Compares it with the traditional contact type measuring technique to have the following merit: 1) on the bad environmental condition or the dynamic work condition, it can measure the components. 2) It dose not need with the parts list surface contact and cannot create the damage to the components surface. 3) The measuring speed is quick and suits in the real-time inspection. Therefore, people have launched extensive research on the visual measuring technology. The system calibration technology is the premise and the foundation in the visual measuring system. Its precision is directly affecting the measurement system measuring accuracy. In this paper, we proposed a measuring simple based on the pixel equivalent, which has the characteristic of fast and simple etc.In the existing vision technology measuring technique, the measurement process contains several stages, in which most importantly are establishment relations between the useful information in the image and the spare part actual size. Therefore, this paper based on black and white check board and automatically, using sub-pixel vertex examination technology examine the precise vertex coordinates, then established the calibration model and compared the method based on the coordinates measurement and based on the pixel equivalent measurement .Finally, detecting the edge of the components in the image based on the method of Sub-pixel, and using this measurement system measures the standard gage block for confirming the accuracy of calibrated system .1.The purpose of calibrating the measurement system is establishment relations between the useful information in the image and the spare part actual size. In this paper, through the analysis and comparison, proposed the calibration method based on the pixel equivalent, which determined a pixel in the image correspondence actual object physical size. Using the model to calibrate the system is very easy, the calibration result is quite precise, and it can avoid expanding the error that caused by the calculated the object world coordinates. Therefore the model is becoming to calibrate the measure system fieldwork.2.The purpose of feature extraction is to locate the features from the images. For the black and white check board is how to get the vertex coordinates in the images. In this paper, examining the vertex coordinates based on the Harris Sub-Pixel vertex examination theory, this method increased the examination precision, then increased the calibration system precision. The edge detecting is a very important process of the measure. Though analyzing the Gray moment operators and the fitting method based on the step edge model, discovered that the fitting method is more accurately and insensitive to the noise.3.The measurement precision of the system that calibrated by the method in this paper is analyzing though measuring different size of standard gage block.This paper is composed of six chapters. The first chapter introduced the background and significance of this work, and summarized the existing calibration method and development condition, finally elaborated the main content of this paper. The second chapter has made a detailed introduction for the system calibration theory and basic concept, and made the exhaustive analysis to the classics calibration method. The third chapter has compared the calibration method based on coordinates measurement and the pixel equivalent, and analyzed them good and bad point, finally, found the appropriate measurement system calibration method. The fourth chapter has introduced the Harris vertex extraction theory and examined the calibration board that used in this paper, then based on the actual components edge type, analyzed and compared two edge detecting method ,Gray moment operators and the fitting method.The five chapter has confirmed the precision of the calibration proposed in this paper though the experiment. The six chapter is the part of conclusion, it summarize the production of this and point out the insufficiencies of this paper at last.