Three-Channel Parallel 3D Topography Measurement of High-Reflective Objects

Literature Overview

The paper by Han Hexiang, Gao Nan, Zhang Guofeng, Guo Tong, Bai Yajing, Ni Yubo, Meng Zhaozong, and Zhang Zonghua (2024, Acta Photonica Sinica, Vol. 53, No. 7, pp. 181-193) proposes a novel three-channel parallel stripe projection method for three-dimensional topography measurement of high-reflective objects. The method addresses the well-known problem of camera overexposure when measuring highly reflective surfaces using conventional structured light techniques, and achieves measurement error reduction to 61.2% of the conventional method while requiring only one additional pre-projection image compared to traditional multi-exposure approaches.

Problem Statement and Technical Challenges

Conventional stripe projection (structured light) measurement systems face significant challenges when applied to high-reflective objects such as polished metal surfaces, coated components, and glossy industrial parts:

• Overexposure: High reflectivity causes the camera sensor to saturate at certain pixel locations, resulting in loss of stripe information and measurement failure in those regions.
• Limited dynamic range: A single exposure setting cannot simultaneously capture both the bright highlights and the darker shadow regions of a reflective surface.
• Inefficient multi-exposure methods: Traditional approaches that acquire multiple images at different exposure times require a large number of frames, increasing measurement time and introducing sensitivity to object or system motion between frames.

Method Description and Technical Principles

The proposed method leverages the differential spectral response of a color camera across the red, green, and blue channels. The key technical steps are:

1. Intensity ratio calculation: The optimal projection intensity ratio for the red, green, and blue channels is determined based on the light intensity histogram distribution of the target object.
2. Parallel RGB stripe projection: Corresponding red, green, and blue stripes are projected simultaneously using the calculated intensity ratios.
3. Channel separation and mask generation: The color camera captures the projected pattern, and the different spectral responses of the three channels allow separation of the corresponding stripe images. For each channel, pixels that are not saturated and have the highest modulation are selected to generate a channel-specific image mask.
4. High dynamic range (HDR) stripe image synthesis: The three channel-specific masks and stripe images are combined to synthesize a high dynamic range stripe image that covers the full intensity range of the reflective surface.
5. Phase extraction and system calibration: Standard phase unwrapping and system calibration procedures are applied to recover the three-dimensional topography.

Performance Comparison

The reduction in measurement error to 61.2% of the traditional method represents a significant improvement, particularly for applications requiring high-precision dimensional measurement of reflective surfaces.

Technical Interpretation

The fundamental insight behind this method is that the three color channels of a standard color camera inherently possess different sensitivities and dynamic ranges. By projecting stripes of different colors with carefully calibrated intensity ratios, the method exploits the spectral selectivity of the camera to capture complementary information from the same scene in a single exposure. The mask generation step ensures that only the most reliable pixel data from each channel contributes to the final HDR image, effectively rejecting saturated or low-modulation pixels.

This approach is particularly advantageous for industrial metrology applications where measurement speed and robustness are critical. The single-capture nature of the method eliminates motion artifacts that can occur between frames in multi-exposure approaches, making it suitable for measuring objects on production lines where vibration and thermal drift are common concerns.

Engineering Practice Implications

For pipe and pipe fitting manufacturing, this measurement technique has direct applications in:

• Surface roughness measurement: Polished or coated pipe surfaces can be accurately characterized without the overexposure issues that plague conventional structured light systems.
• Dimensional inspection: Three-dimensional topography data can be used to verify the geometric accuracy of formed pipe fittings, including wall thickness distribution, ovality, and surface profile deviations.
• Weld quality assessment: The technique can be applied to inspect weld bead geometry, including weld height, width, and profile, which are critical quality indicators for pipe welding operations.

The method's ability to reduce measurement error by nearly 40% compared to traditional approaches makes it suitable for high-precision applications such as aerospace pipe fitting inspection, where dimensional tolerances are extremely tight.

Reflections and Study Insights

This paper represents a significant advance in structured light metrology for challenging surface types. The clever exploitation of color channel spectral differences to achieve HDR imaging in a single capture is an elegant solution to a long-standing problem in optical metrology. However, the method's effectiveness depends on the accuracy of the initial intensity ratio calculation, which requires a pre-projection image to determine the object's light intensity distribution. For highly variable surfaces, the pre-projection step may need to be repeated to ensure accurate intensity ratio estimation.

Additionally, the method assumes that the three color channels have sufficiently different sensitivities to provide complementary information. For surfaces with strong spectral selectivity (e.g., colored coatings), the channel separation may be less effective, potentially reducing the HDR improvement. Engineers applying this technique should validate its performance on their specific target surfaces before deploying it in production metrology systems. This literature provides a valuable reference for engineers seeking to improve three-dimensional measurement accuracy for reflective industrial surfaces, and the method's single-capture advantage makes it particularly attractive for online inspection applications.