Modeling the Anisotropic Reflectance of a Surface with Microstructure Engineered to Obtain Visible Contrast After Rotation

Abstract

Engineering of surface structure to obtain specific anisotropic reflectance properties has interesting applications in large scale production of plastic items. In recent work, surface structure has been engineered to obtain visible reflectance contrast when observing a surface before and after rotating it 90 degrees around its normal axis. We build an analytic anisotropic reflectance model based on the microstructure engineered to obtain such contrast. Using our model to render synthetic images, we predict the above mentioned contrasts and compare our predictions with the measurements reported in previous work. The benefit of an analytical model like the one we provide is its potential to be used in computer vision for estimating the quality of a surface sample. The quality of a sample is indicated by the resemblance of camera-based contrast measurements with contrasts predicted for an idealized surface structure. Our predictive model is also useful in optimization of the microstructure configuration, where the objective for example could be to maximize reflectance contrast.

Microstructure Control in 3D Printing with Digital Light Processing

“…In this example, we used the greyscale pattern for the last layer of the printing process only. We used the 2D sinusoid to generate the patterns in the main diagonal of Figure , with parameters µ m and µ m, respectively, µ m and µ m. In the antidiagonal, we used ridged patterns [LFD*17] with an inclination of 10° and pitch length of µ m. The ridges of the patterns in these two smileys have orthogonal orientations. The QR code in Figure is another example of a surface with orthogonal ridged structures, but this was printed using the 2 × magnifying lens.…”

“…We print anisotropic reflectance properties using a 2D sinusoidal function with different frequencies along the two axes, or a sequence of parallel ridges, as described by Luongo et al . [LFD*17], see Figure . These patterns are useful for producing anisotropic reflectance contrast (smileys and QR code in Figure ).…”

“…For the ridged structure in Figure , we use the model presented by Luongo et al . [LFD*17]. For the anisotropic sinusoidal patterns, we derived a new model, which is described in Appendix B.…”

Design of rough microgeometries for numerical simulation of material appearance

“…Many authors have proposed to generate heightfields from height probability distributions and autocorrelation [26,27], but they cannot be extended to slope distributions [2]. C 0 continuous surface have been generated by several authors from Gaussian slope probability distributions [4], or continuous C −1 surface (i.e., with height discontinuities) from non-Gaussian slope probability distributions [28][29][30]. Weyrich et al [31,32] propose an approach suitable for non-Gaussian surfaces, but it still does not ensure surface continuity and it is not designed to reproduce smooth distributions [33].…”