Research On 3D Measurement Method Based On Line Structured Light

With the iteration of measurement technology,measurement technique which uses vision is becoming more popular.Detection technology based on 2D vision plays an important role in the traditional visual field,but it is often difficult to meet the detection requirements in the face of vertical dimension,so 3D vision detection technology arises at the right moment.Among such technique,structured light 3D measurement method does not get in touch with surface of object directly.Because it has high real-time detecting ability and high accuracy,so it has been widely used in measurement area and detection field.In this paper,study on the method about 3D measurement which based on structured line light,the laser beam through the lens module in the visual sensor is formed on the light stripe image,through the light stripe image processing and calibration of the whole measurement system,can achieve two-dimensional laser stripe image and the spatial three-dimensional contour point cloud data of conversion.This paper studies the key technologies in the 3d measurement method of line structured light,mainly including the following aspects:(1)Study the principle of scheimpflug imaging and the measurement principle of line structured light 3D measurement system,and establish 3D line laser scanning measurement system,set up line structured light 3D measurement system,establish a 2D image pixel coordinates to 3D space coordinate transformation model between,as long as the model parameters are solved out,let the measurement system and the object being measured at a certain rate of relative movement,It can realize scanning reconstruction of object's 3D contour.(2)To study the parameter calibration method of imaging system based on scheimpflug imaging characteristics,it is necessary to integrate a layer tilt model on the basis of the traditional calibration model to calibrate the camera's focal length,main point coordinates,scheimpflug Angle and rotation vector.To study the distortion of image,a calibration method of distortion center and distortion coefficient is designed based on collinearity of corner points.Just need one checkerboard image,we can carry out the calibration of distortion parameters.To study the calibration of the whole system,an adaptive segmenting node calibration method is proposed.The whole calibration area is divided into three parts for calibration,which can reduce the problem of low calibration accuracy caused by system error to a certain extent.By substituting calibrated parameters into the whole measurement system,accurate contour point cloud data can be obtained.(3)The center of gravity extraction algorithm of line structured light fringe is studied.Firstly,the preprocessing algorithm of light fringe image is studied according to the imaging characteristics of laser fringe.The location of the center point of the light fringe was obtained by using the extremum method at the pixel level,and then linear interpolation was carried out and the gray center method to carry out second result.The result in experiment shows that the precision of the center of the light fringe extraction algorithm designed in papers results is better than traditional center of mass method.(4)Verify the measurement accuracy of the whole system.In this paper,the standard measuring block is used to verify the absolute measurement accuracy and repeated measurement accuracy.Firstly,using the least square method to fit base plane after dealing with cloud data,and then calculating the distance between base plane and the upper plane of standard block.Experimental results show that the maximum absolute measurement error of the whole measurement system is 5um,the average measurement accuracy is 3.4um,and the repeated measurement accuracy is 0.35 um.This proves that this measurement system is feasible and robust.