Micrometer-level 3D measurement techniques in complex scenes based on stripe-structured light and photometric stereo

Zhao Song, zhan song

Abstract:

Compared with existing depth cameras, such as RGB-D, RealSense and Kinect,
stripe-based structured light (SL) has the potential for micrometer-level 3D measurement; this
can be attributed to its higher coding capacity. While surface texture, high-reflective region,
and occlusion remain some of the main sources leading to degraded reconstruction quality in
complex objects, methods that are only based on SL cannot completely solve such problems
in complex object reconstruction. In this paper, we developed an advanced fusion strategy for
the reconstruction of complex objects in micrometer-level 3D measurement. This includes
solving the above-mentioned inherent problems of a stripe-based SL system with the aid of
photometric stereo (PS). Firstly, to improve the robustness of decoding and eliminate the effects
of noise and occlusion on stripe detection, a novel scene-adaptive decoding algorithm based
on a binary tree was proposed. Further, a robust and practical calibration method for area
light sources in the PS system, which utilizes the absolute depth information from SL system,
was introduced. A piecewise integration algorithm, which is based on a subregion divided by
Gray code, was proposed by combining the depth values from SL with the normal information
from PS. Remarkably, this method eliminates the effects of surface texture and high-reflective
region on the reconstruction quality and improves the resolution to camera-level resolution.
In experimental parts, a regular cylinder was reconstructed to demonstrate micrometer-level
measurement accuracy and resolution enhancement by the proposed method. Then, improvement
of the reconstruction accuracy for objects with surface texture was validated with a regular
pyramid that had textures on it and a white paper with characters printed on it. Lastly, a complex
object containing multiple phenomena was reconstructed with the newly proposed method to
show its effectiveness for micrometer-level 3D measurement in complex objects. Evaluation of
our proposed method shows the improvement of the proposed method on the existing methods
being used for micrometer-level 3D measurement in complex objects.