The development of a low-cost photometric-stereo-based material scanner

Area of Science:

• Computer Graphics
• Digital Imaging
• Material Science

Background:

• Accurate texture map capture is crucial for digital prototyping and spatial augmented reality (SAR).
• Existing methods can be costly and complex, posing challenges for accessibility.
• Photometric stereo offers a potential solution for efficient material digitization.

Purpose of the Study:

• To present an open-source, affordable, and user-friendly material scanning system.
• To enable efficient capture of albedo and normal maps for physically based rendering (PBR).
• To evaluate the system's performance against professional digitization methods.

Main Methods:

• Developed an affordable system using a consumer-grade camera, MR16 LED lighting, and Arduino automation.
• Implemented a hybrid calibration workflow for accurate color reproduction.
• Utilized a streamlined Substance 3D Designer pipeline for texture map generation.
• Employed CIEDE2000 (ΔE00) for albedo map accuracy and visual/directional analysis for normal maps.

Main Results:

• The system captures material data up to A4 size within 15 seconds.
• Generated albedo and normal maps are compatible with physically based rendering (PBR).
• Qualitative comparisons with professional references show perceptually consistent textures.
• Color accuracy assessed using CIEDE2000 (ΔE00) demonstrates reliable performance.

Conclusions:

• The proposed system offers a low-cost, customizable solution for material digitization.
• It provides perceptually consistent textures suitable for real-time rendering applications.
• The open-source nature promotes accessibility and further development in digital prototyping and SAR.