Xiaojie Zhang

This Letter presents a ray-resampling method for direct geometric distortion correction based on densely calibrated pixel-to-ray mappings. Instead of relying on system-specific distortion models, calibrated rays are intersected with an object-space reference plane and resampled onto a regular grid, allowing the same correction workflow to accommodate substantially different imaging geometries. The method is further extended to light-field imaging through sub-aperture ray reconstruction and resampling. Experiments on wide-angle, fisheye, and non-central catadioptric systems achieve overall linearity errors of 0.0202, 0.0337, and 0.0557 px, respectively. In structured light-field 3D reconstruction, the standard deviation of plane-fitting error decreases from 1.8386 mm to 0.1709 mm, demonstrating accurate geometric correction and improved downstream 3D measurement.

Binary fringe projection holds significant promise for three-dimensional structured illumination microscopy (SIM) due to its high efficiency and precision. However, its performance is limited by fringe edge blurring and localized over- or underexposure on highly reflective surfaces, leading to demodulation failure and incomplete measurement. This paper proposes a high dynamic range SIM based on multi-channel fusion and dual-phase binary fringe (MFDBF-SIM). By introducing a phase shift between axially adjacent binary fringes, the zero-modulation regions resulting from edge blur are distributed alternately between odd and even layers, thereby avoiding local information loss. Furthermore, differential intensity responses across RGB channels are utilized to adaptively select the optimal modulation curve, enabling high dynamic range reconstruction. A horizontal correction model and a scanning-step-based calibration model are developed to compensate for horizontal error caused by Bayer interpolation and axial focal-plane offset induced by lens longitudinal chromatic aberration, respectively. Experimental results on silicon wafer and printed circuit board samples demonstrate that MFDBF-SIM improves reconstruction integrity by 121.5% and 224.6% compared to the single-channel methods using only the green or blue channel.

In industrial measurement, the high reflectivity of metallic surfaces and the complex, time-varying ambient light interference (TVALI) are two primary factors that limit the accuracy of fringe projection profilometry (FPP). The former often causes local saturation and reduces fringe contrast, whereas the latter introduces a time-varying background component that compromises the accuracy of phase retrieval. To address these challenges, we propose a polarization-modulated illumination recovery model (P-MIRM) based on the physical mechanism of Fresnel reflection on metallic surfaces. By taking advantage of the strong polarization-preserving property of linearly polarized projected fringes on metallic surfaces, we adopt the response differences among the four channels of a polarization camera to achieve pixel-wise decoupling of time-varying ambient illumination from the modulated signals using a single image. Compared with conventional schemes that rely on mechanical rotation, the proposed method requires no additional operations, which substantially simplifies system calibration and improves measurement efficiency. Experimental results on the metal standard piece show the proposed method reduces the standard deviation of plane-fitting residuals and the error of adjacent step height differences by approximately 73% and 47%, respectively, demonstrating markedly improved robustness under unstable illumination.

Structured light field (SLF) 3D measurement integrating micro-electro-mechanical system (MEMS) projection with unfocused plenoptic cameras (UPC) offers a promising solution for large depth of field (DOF) metrology. However, the high coherence of laser illumination and small UPC sub-apertures generates severe speckle and low signal-to-noise ratio (SNR), degrading reconstruction accuracy. To alleviate these issues, this paper proposes an EPI-guided adaptive refocusing model (EGARM). The model employs a phase-encoded epipolar plane image (EPI) analysis to extract phase features and adaptively estimate a depth-related slope factor. This factor subsequently guides digital refocusing to aggregate multiple sub-aperture views from the 4D light field. By averaging speckle patterns across viewpoints, the method synthesizes sharp images while effectively suppressing speckle. Additionally, a refocused ray calibration and global geometric refinement framework is introduced to impose global geometric constraints and help limit the degradation of reconstruction accuracy across the measurement volume. Experimental results demonstrate stable 3D reconstruction over an extended depth range of approximately 400-1856 mm, during which the effective field of view (FOV) expands from 120 mm × 200 mm to 580 mm × 780 mm, with reconstruction accuracy consistently better than 0.45 mm.

Fringe projection profilometry (FPP) enables high-precision 3D shape measurement and is increasingly deployed with biaxial micro-electro-mechanical system (MEMS) laser-scanning projectors for large-range projection. However, the fast and slow axes of a biaxial MEMS mirror exhibit coupled and nonlinear motion, causing the actual scanning trajectory to deviate from an ideal separable sweep. Consequently, points sharing the same phase no longer lie on straight lines or planar isophase surfaces, but instead follow curved spatial loci, making accurate calibration challenging. This work introduces a phase-angle-ray model (PARM) that establishes a bivariate polynomial mapping between the phase and MEMS deflection angles, from which projected rays are analytically derived. Two calibration frameworks based on pinhole and ray-based imaging models are developed by integrating geometric constraints from a standard plane and multiple standard spheres. Experiments demonstrate that the proposed method achieves high calibration accuracy, improved metric consistency, and precise 3D reconstruction.

Structured illumination microscopy for optical sectioning is widely used for micro- and nanoscale surface topography measurement, but the conventional phase-shifting approach requires multiple image acquisitions, leading to a trade-off between measurement speed and accuracy. To overcome this limitation, we propose a complementary binary fringe difference method for structured illumination microscopy to obtain optical sectioning at each layer. Theoretical derivation and simulation experiments demonstrate that the modulation signal obtained through complementary binary fringe difference exhibits an axial intensity profile approximating a symmetric Bessel function curve. This enables the determination of depth positions corresponding to peak intensities via Gaussian function fitting, thereby realizing surface shape measurement. Experimental results on a 1.8 µm step-height sample demonstrated the effectiveness of the proposed method, achieving a root-mean-square error of 29 nm.

Fringe projection profilometry (FPP) is widely used in industrial metrology and 3D scanning. However, relying on traditional lens-based imaging mechanism, existing FPP systems face challenges in achieving high-precision measurement over large depth range. To overcome this limitation, this paper presents an approach integrating micro-electro-mechanical system (MEMS) projection with light field imaging (LFI). To accurately characterize the projection-imaging relationship across a large depth of field (DOF), an extended depth ray fusion model (EDRFM) is proposed, utilizing ray tracing to unify the representation of both projection and imaging processes. Furthermore, phase consistency ensures spatial alignment across multiple viewpoints, and a fusion method incorporating the angle between projection-camera rays and phase modulation quality is introduced to reduce errors caused by view variations. Experimental results demonstrate that the proposed method achieves a measurement accuracy of 0.12 mm within a 650 mm depth range.

Fringe projection profilometry (FPP) faces significant challenges regarding calibration difficulty and stitching error accumulation when operating across scenes ranging from tens to hundreds of meters. This Letter presents a calibration-free 3D measurement method by integrating a binocular vision of a FPP scanner with a wide field-of-view (FoV) vision that constructs global benchmarks to unify local 3D scanning and global 3D stitching, which is adaptable to arbitrarily large-scale scenes. A posterior global optimization model is then established to determine the reconstruction parameters and stitching poses simultaneously at each scanning node with adaptively distributed benchmarks. Consequently, the integrated vision measurement system not only eliminates the large-scale pre-calibration and stitching error accumulation but also overcomes system structural instability during moving measurement. With the proposed method, we achieved 3D measurements with an accuracy of 0.25 mm and a density of 0.5 mm for over 50-m-long scenes.

The binocular structured light 3D measurement system is widely used in situ industrial inspection and shape measurement, where the system structure is generally unstable due to mechanical loosening or environmental disturbance. Timely corrections to the changing structural parameters thus is an essential task for online high-accuracy measurement, which is difficult for traditional unidirectional fringe projection methods to self-correct the structural change. To this end, we propose an online self-correction method based on the investigation that orthogonal fringe projection can intrinsically relax the constraint on the epipolar geometry relationship and provide bidirectional phases for accurate corresponding point searching. Since orthogonal fringe projection may sacrifice the measurement efficiency, we further design a searching strategy by locally unwrapping one directional phase to reduce the number of projection patterns. Experimental results demonstrate that the proposed method is effective for online self-correction of unstable system structure to achieve high-accuracy 3D measurement under complex measurement environments.