For Line-Laser sensor products that CCD images are unknown, we present a method for the calibration of Line-Laser sensor measurement system using multi-directional and non-featured planes, and a method for system calibration optimization using multi-angle standard spheres. By building a mathematical model, we convert the line laser sensor measurement data into CMM measurement points. According to the constraint relationships of planes or spheres, the point measured by the Line-Laser sensor and the CMM should conform to the same equation, then we can solve the calibration matrix of the line laser sensor and the coordinate measuring machine by nonlinear optimization. Both simulation analyses and real experiments were conducted. A line laser sensor was used to measure a frosted standard ball with a radius of 12.696 mm. The radius deviation measured by the line laser sensor system and the center deviation of the sphere comparing with the CMM were observed. The experimental results show that the radius deviation of the calibration laser sensor measurement system is less than 0.02mm, and the center distance deviation of the sphere is less than 0.02mm. This method utilizing non-featured planes simplifies the calibration equipment and can reduce the fitting error when using standard ball from multiple angles for calibration. This method is different from the method of calibrating the single direction of the laser sensor. It can simultaneously calibrate the rotation matrix and translation matrix of the two-dimensional line laser sensor to the coordinate measuring machine, and optimize the global optimal calibration parameters.
Laser scanning is widely used in on-line industrial 3D inspection, cultural heritage conservation and reverse engineering.
However, in the traditional laser scanning, the most widely-used approach is based on the projection of a single directional
laser stripe over an object. Because the width of the laser stripe is physically difficult to compress enough to be fine at the
edge of the object, the traditional measurement method is not accurate for edge measurements. This paper proposes an
edge sensitive 3D measurement system which is fast and accurate, using two directional laser stripes scanning with a laser
projector. Scanning metal edge steps and complex surface edge with this system only require a single scanning to perform
3D reconstruction. So this scanning method has the advantages of high efficiency, high speed and edges with high precision.
Line structure light measurement needs accurate mechanical movement device and high -frame-rate camera, which is
difficult to realize. We propose a high-speed full-field profilometry to solve these difficult ies, using coded laser strips
projected by a MEMS scanning mirror. The mirror could take place of the mechanical movement device with its high
speed and accurate. Besides, a method with gray code and color code is used to decrease the frames number of projection,
retaining the advantage of line structure light measurement. In the experiment, we use a laser MEMS scanner and two
color cameras. The laser MEMS scanner projects coded stripes, with two color cameras collecting the modulated pattern
on the measured object. The color cameras compose a stereo vision system so that the three-dimensional data is
reconstructed according to triangulation.