Rapid Three-Dimensional Measurement Using Three-Channel Binary Stripe Defocused Projection

Literature Overview

This paper by Zhang Xin, Gai Shaoyan, and Da Feipeng, published in Laser & Optoelectronics Progress (2020, Vol. 57, No. 23), presents a structured-light-based three-dimensional measurement method that encodes phase-shifted stripe patterns into three color channels and projects them sequentially within a single camera frame time. The work addresses a longstanding challenge in industrial metrology: achieving both high speed and high accuracy when measuring complex surfaces, particularly those with varying reflectivity or color. Funded by the National Natural Science Foundation of China (Grants 51475092 and 61462072) and the Jiangsu Provincial Natural Science Foundation (Grant BK20181269), the research was conducted at Southeast University's School of Automation and the Key Laboratory of Measurement and Control of Complex Engineering Systems under the Ministry of Education.

Core Technical Principles

The method relies on three key innovations that distinguish it from conventional color-coded structured light systems. First, instead of projecting continuous phase-shifted patterns, the authors employ binary stripe patterns separated across three color channels. At the projector, three independent binary digital images serve as inputs, and the defocused stripes of all three channels are projected sequentially within one camera frame time. Second, a color camera captures the fused three-channel stripe image, which is then decoded through a channel-decoupling procedure. Third, a black-and-white camera is used to calibrate the color-channel aliasing effect inherent in the measurement system, thereby compensating for the spectral cross-talk that typically degrades reconstruction accuracy.

The following table summarizes the key technical parameters and advantages of the proposed method:

From a metallurgical and pipeline engineering perspective, the significance of this work lies in its potential application to dimensional metrology of pipe fittings. Elbows, tees, and reducers manufactured by hot bending, cold bending, or forging often exhibit geometric deviations that affect their mechanical performance and pressure-containing capacity. Traditional contact measurement methods are slow and may not capture the full three-dimensional surface topology. Optical structured light methods offer non-contact, full-field measurement capability, but conventional approaches struggle with the dark, metallic, or painted surfaces common in industrial pipe components. The proposed method's ability to handle colored and dynamic objects makes it particularly attractive for in-line inspection of pipe fittings during manufacturing or during field inspection of installed piping systems.

Interpretation of Technical Points

The use of binary stripes rather than continuous phase-shifted patterns represents a fundamental shift in the measurement paradigm. Binary patterns are more robust to noise and partial occlusion because each pixel is either fully illuminated or fully dark, creating a higher signal-to-noise ratio in the captured image. The defocused projection technique further enhances this robustness by creating smoother, more continuous stripe patterns that are less susceptible to surface texture variations. This is particularly important when measuring the inner surface of pipes or the curved surfaces of elbows, where surface finish can vary significantly due to manufacturing processes such as hot bending, cold bending, or welding.

The channel-decoupling procedure is critical to the method's accuracy. In a three-channel color system, the red, green, and blue channels are not perfectly independent; spectral overlap and sensor cross-talk cause aliasing effects that can distort the measured phase and, consequently, the reconstructed geometry. By using a monochrome camera to characterize this aliasing effect during system calibration, the authors effectively separate the three channel contributions from the fused color image. This calibration step is analogous to the correction of systematic errors in dimensional metrology instruments, such as the thermal expansion correction applied to coordinate measuring machines or the probe calibration performed in ultrasonic testing.

Application to Pipeline and Fitting Inspection

In the context of steel pipe and fitting quality control, three-dimensional surface measurement is essential for several applications. First, the geometric accuracy of bent elbows must be verified against standards such as ASME B16.9, which specifies tolerances for bend radius, wall thickness reduction, and ovality. Second, the weld geometry of butt-weld fittings and socket-weld fittings must be inspected for proper reinforcement, leg length, and profile continuity. Third, during pipeline integrity assessment, the three-dimensional geometry of in-service elbows and bends is critical for stress analysis and remaining life evaluation, particularly in regions subject to bending deformation, corrosion, or mechanical damage.

The proposed method's speed advantage is especially relevant for in-line inspection scenarios where the inspection tool must traverse the pipeline at operational speeds while capturing high-resolution geometric data. Conventional magnetic flux leakage (MFL) tools can detect geometric anomalies such as dents and ovality but do not provide full three-dimensional surface reconstruction. The combination of structured light metrology with pipeline inspection technology could enable a new class of inspection tools capable of characterizing the complete three-dimensional geometry of elbows, tees, and reducers during in-line inspection runs.

Key Questions and Reflections

Several questions arise from a practical engineering standpoint. First, the method relies on structured light projection, which requires line-of-sight access to the measurement surface. In pipeline inspection, the internal surface of a pipe is accessible only through a confined bore, and the projection geometry must be carefully designed to accommodate the cylindrical geometry and limited working distance. Second, the surface finish of steel pipes and fittings can range from rough mill finish to polished surfaces, and the reflectivity and specular characteristics of these surfaces can affect the quality of the captured stripe patterns. Defocused projection helps mitigate this issue, but the effectiveness on highly reflective stainless steel surfaces remains to be verified. Third, the method's accuracy must be validated against established dimensional metrology standards, such as those specified in ASME B16.9 for butt-weld fittings or GB/T 12459 for steel pipe fittings.

The work demonstrates that structured light metrology can be made sufficiently fast and accurate for industrial applications when the stripe encoding scheme and channel calibration are carefully designed. For pipeline and fitting engineers, this opens the possibility of integrating high-speed three-dimensional measurement into manufacturing quality control and field inspection workflows, complementing traditional methods such as ultrasonic testing, radiographic testing, and dimensional gauging. The ability to measure colored and dynamic objects is particularly valuable for in-service inspection, where pipe surfaces may be coated, painted, or covered with corrosion products.

Study Insights and Implications

This paper contributes to the broader field of industrial optical metrology by demonstrating that the combination of binary stripe encoding, defocused projection, and channel-decoupling calibration can overcome the speed-accuracy trade-off that has limited the adoption of structured light methods in manufacturing environments. For engineers working in pipe and fitting quality control, the key takeaway is that advanced optical measurement techniques are maturing rapidly and may soon become practical tools for in-line inspection and manufacturing quality assurance. The method's applicability to colored and dynamic objects suggests that it could be adapted for use in automated inspection cells where pipe fittings move through the measurement station at production speeds. Future work should focus on adapting the projection geometry for cylindrical pipe surfaces, validating the method against established dimensional standards, and integrating it with existing pipeline inspection tool trains to provide complementary geometric data alongside conventional defect detection signals.