Research On Key Techniques Of Line-Structured Light Measurement System

With the development of machine vision related technology,visual measurement technology has become an important means of contactless measurement technology,in which line structure optical measurement technology is widely used in industrial production because of its advantages of high precision and fast speed.With the development of the industrial field and the advancement of science and technology,the requirements for measurement technology are getting higher and higher,and higher demand for measurement accuracy and reliability of line structured light measurement system are put forward.Therefore,it is urgent to carry out line structure light vision three-dimensional measurement key technology research.In this paper,the actual requirements for line structure light vision measurement.The optical path system of line structured light vision sensor is designed.The structural parameter error of Line Structured light vision sensor is analyzed.The traditional calibration method and direct calibration method is studied.The feature extraction method of optical pattern for high light surface structure is studied.The line structure light vision sensor is designed based on the direct laser triangle method,and the optical path system of the vision sensor is designed,analyzed and calculated,and the appropriate optical path system parameters and components are selected.According to the selected optical device,the measurement model of linear structure optical vision sensor is completed and the corresponding mathematical model is established.Based on the analysis of the imaging characteristics of the sensor,combined with the theory of measurement error,the constraint equations whose structural parameters need to be satisfied when designing the sensor are obtained,and the best results are obtained to solve the requirements of the performance index of the error and resolution.The calibration of linear structure optical vision sensor mainly has traditional calibration method and direct calibration method.The traditional calibration method mainly includes two parts,the calibration of the camera and the calibration of the structural parameters of the optical vision sensor of the line structure.The camera calibration adopts the classical planar template calibration method,and the structural parameters of the optical vision sensor of the line structure are obtained by fitting the feature points on the optical plane of the line structure,and the sensor calibration is completed.Aiming at the calibration problem of optical vision sensor of linear structure,the number of images by the camera calibration,the extraction method of the central line of the structure light pattern and the selection quantity of the structural optical plane feature points three key influencing factors are analyzed,in order to improve the calibration accuracy of the sensor.The optimal parameters in the calibration process are determined by comparative experimental study.In the direct calibration method research,a calibration target consisting of high-precision glass stripe plate and precision translation stage is designed to extract the sub-pixel coordinate values of the target image feature points and their corresponding spatial three-dimensional coordinate values.The support vector machine and BP neural network method are used to establish the mapping relationship between the image coordinates of the feature points and the spatial three-dimensional coordinates,and the direct calibration of the line structure light vision sensor is realized.The experimental results show that the direct calibration method can meet the requirements of accuracy and speed.Aiming at the extraction of the center line of the high-light surface structure of the strong reflective material,a light-grain center extraction algorithm based on the Gaussian curve fitting method of the peak boundary is proposed.The accuracy and algorithm accuracy of the algorithm are compared and analyzed.Finally,the measurement experiments are carried out to verify the proposed calibration method of linear structure optical vision sensor and the effectiveness of the surface image processing algorithm for strong reflective materials,and the experimental results show that the optical vision measurement method of linear structure in this paper has good stability and higher measurement accuracy.