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

Andrea Luongo, Viggo Falster, Mads Brix Doest, Dongya Li, Francesco Regi, Yang ZhangWorkshops @ ICCV 2017 · DOI

Introduction

In modern high-end manufacturing and product design, "appearance printing" has emerged as a sophisticated technique for creating dynamic visual effects. By engineering specific microstructures onto surfaces, designers can produce logos, patterns, or security features that appear or disappear depending on the viewing angle or the object's orientation. Despite its growing popularity, the development of these surfaces has historically relied on trial-and-error experimental methods.

Researchers at the Technical University of Denmark have addressed this lack of theoretical foundation by developing an analytical Bidirectional Reflectance Distribution Function (BRDF). This model predicts how light interacts with microscopic parallel ridges, providing a mathematical framework for optimizing contrast and enabling automated quality inspection in manufacturing.

Methodology

The core of the research involves building an analytical BRDF based on microfacet theory, specifically tailored for surfaces with engineered micro-ridges. The surface is treated as a "macrosurface" composed of a "microsurface" of parallel ridges.

Geometric Modeling

The model focuses on ridges with a pitch typically ranging from 50 to 150 $\mu$m. At this scale, the authors demonstrate that geometrical optics are sufficient to predict visible contrast, avoiding the need for more computationally expensive wave-interference models.

Geometric parameters of the engineered micro-ridges, including pitch and ridge angle ($\theta_r$).

The Geometrical Attenuation Function ($G$)

A significant contribution of this work is the derivation of a Geometrical Attenuation Function ($G$) specifically for ridge geometries. This function accounts for shadowing (where the ridge blocks light from reaching parts of the surface) and masking (where the ridge blocks the camera's view of parts of the surface).

Microfacet Distribution and Scattering

To account for real-world manufacturing imperfections, the model employs an anisotropic Beckmann distribution ($D$). This statistical approach represents the microscopic roughness and irregularities found on the slopes of the ridges. Furthermore, the final model integrates both direct specular reflection and subsurface scattering—light that penetrates the material before being reflected back—making it highly effective for modeling colored plastic surfaces.

Results and Discussion

The analytical model was validated against experimental measurements involving physical samples with varying ridge angles ($\theta_r$), camera tilt angles ($\theta_c$), and azimuthal rotations ($\phi_s$).

Trend Alignment and Rotation Contrast

The model successfully captured the primary "anisotropic" effect: a significant change in contrast when the surface is rotated by 90 degrees. It also accurately predicted that maximum contrast occurs when the camera is not tilted ($\theta_c = 0^\circ$), with contrast decreasing as the tilt angle increases.

Simulated checkerboard patterns demonstrating the anisotropic reflectance and contrast change upon rotation.

Virtual Prototyping

By utilizing the BRDF to render synthetic images, such as the checkerboard patterns shown above, the researchers demonstrated the capability for virtual prototyping. This allows designers to simulate and evaluate the appearance of engineered surfaces before physical production, significantly reducing development time and costs.

The "Noise" Gap in Manufacturing

An interesting finding emerged from the comparison between the model and physical experiments. While the model predicted that a shallow $5^\circ$ ridge angle would yield the highest contrast, physical samples performed best at $10^\circ$. The authors hypothesize that manufacturing artifacts, such as tiny burrs or material residues from the milling process, create "noise" that obscures the signal of very shallow ridges. As the ridge angle increases, the intended geometric signal becomes strong enough to overcome this manufacturing noise.

Conclusion

This research provides a robust analytical foundation for the design and inspection of engineered surfaces. By bridging the gap between micro-scale geometry and macro-scale appearance, the model enables more precise control over visual effects in product design. Future work will focus on explicitly modeling manufacturing defects to use this BRDF as a "gold standard" for automated visual quality inspection, comparing manufactured parts against ideal model predictions.

Summary updated

BibTeX
@inproceedings{luongo2017modeling,
  author = {Andrea Luongo and Viggo Falster and Mads Brix Doest and Dongya Li and Francesco Regi and Yang Zhang},
  title = {{Modeling the Anisotropic Reflectance of a Surface with Microstructure Engineered to Obtain Visible Contrast After Rotation}},
  booktitle = {Workshops @ ICCV},
  year = {2017},
  doi = {10.1109/iccvw.2017.27},
}
Figure