Bilayer Synchronous Measuring Method Of Curved Screen Based On Line-structured Light Scanning Sensor

The curved surface screen,which is defined as the cover surface screen of thin transparent material,is generally used on non-flat display devices.It is beloved by the public because it has the advantages of reducing glare and excellent visual effects,so it is widely used in smartphones,smart wear,vehicles,and aerospace engineering.However,because of the poor control stability of the precision of surface screen processing,there are often some problems with curved screen products now,such as inaccurate geometric size,poor planeness and uneven thickness,which seriously affects the normal function of display devices.Therefore,to meet the demand for large-areas curved screen measurement,it is urgent to find a fast,efficient,low-cost and multi-layer synchronous measurement solution.Aiming at solving the problems above,the following research work is systematically carried out in this subject.1.Based on full investigation on the existing curved screen measurement technology,we explore the feasibility of curved screen measurement deeply which is based on the principle of the line-structured light.After fully analyzing the current technical difficulties,we put forward a technical route suitable for the rapid multicontour measurement of the large-area curved screen.2.To solve the problems of little SNR and imprecision of center extraction of a two-dimensional contour image,which comes from the data from a single column scan by using line-structured light,the image enhancement partition and the weighted center of gravity algorithm are used to extract multi-layer contour.The multi-dimensional denoising algorithm is adopted to remove the noise,so the high-precision extraction of the line-structured light outline can be realized.At the same time,a direct calibration improvement model based on sawtooth calibration is proposed,which realizes the transformation of coordinates of object-image space.3.A method of directly compensating the error using measurement data splicing is proposed,which effectively solves the spatial error problem between multi-column scanning measurement data.Through the reasonable downsampling of the data and the timeliness of the improved algorithm,the feasible domain penalty function is used to enhance the optimization constraint,which greatly improves the precision and efficiency of the large-area splicing of the curved screen measurement data.4.An experimental platform for measuring a curved screen with a line-structured light sensor has been designed and built.Based on verifying the accuracy of singlecolumn scanning and the accuracy of region splicing,we have realized the rapid measurement of multi-contour large-area curved mobile phone screens and applied the technical route to the multi-contour measurement of lens devices successfully.