3D Shape Measurement Research Articles

Quasi-pixelwise motion compensation for 4-step phase-shifting profilometry based on a phase error estimation.

Phase-shifting profilometry (PSP) is widely used in 3D shape measurement due to its high accuracy. However, in dynamic scenarios, the motion of objects will introduce phase-shifting errors and result in measurement errors. In this paper, a novel compensation method based on 4-step phase-shifting profilometry is proposed to reduce motion-induced errors when objects undergo uniform or uniformly accelerated motion. We utilize the periodic characteristic of fringe patterns to estimate the phase errors from only four phase-shifting patterns and realize a pixel-wise error compensation. This method can also be applied to non-rigid deforming objects and help restore high-quality texture. Both simulation and experiments demonstrate that the proposed method can effectively improve the measurement accuracy and reduce surface ripples introduced by motion for a standard monocular structured-light system.

Three-dimensional shape and deformation measurement on complex structure parts

Stereo digital image correlation technique (stereo-DIC or 3D-DIC) has been widely used in three-dimensional (3D) shape and deformation measurement due to its high accuracy and flexibility. But it is a tough task for it to deal with complex structure components because of the severe perspective distortion in two views. This paper seeks to resolve this issue using a single-camera system based on DIC-assisted fringe projection profilometry (FPP). A pixel-wise and complete 3D geometry of complex structures can be reconstructed using the robust and efficient Gray-coded method based on a FPP system. And then, DIC is just used to perform the temporal matching and complete full-field pixel-to-pixel tracking. The in- and out-of-plane deformation are obtained at the same time by directly comparing the accurate and complete 3D data of each corresponding pixel. Speckle pattern design and fringe denoising methods are carefully compared and chosen to simultaneously guarantee the measuring accuracy of 3D shape and deformation. Experimental results demonstrate the proposed method is an effective means to achieve full-field 3D shape and deformation measurement on complex parts, such as honeycomb structure and braided composite tube, which are challenging and even impossible for the traditional stereo-DIC method.

Linear Laser Scanning Measurement Method Tracking by a Binocular Vision

The 3D scanning of a freeform structure relies on the laser probe and the localization system. The localization system, determining the effect of the point cloud reconstruction, will generate positioning errors when the laser probe works in complex paths with a fast speed. To reduce the errors, in this paper, a linear laser scanning measurement method is proposed based on binocular vision calibration. A simple and effective eight-point positioning marker attached to the scanner is proposed to complete the positioning and tracking procedure. Based on this, the method of marked point detection based on image moment and the principle of global coordinate system calibration are introduced in detail. According to the invariance principle of space distance, the corresponding points matching method between different coordinate systems is designed. The experimental results show that the binocular vision system can complete localization under different light intensities and complex environments, and that the repeated translation error of the binocular vision system is less than 0.22 mm, while the rotation error is less than 0.15°. The repeated error of the measurement system is less than 0.36 mm, which can meet the requirements of the 3D shape measurement of the complex workpiece.

Motion-induced error reduction for phase-shifting profilometry with phase probability equalization

Phase-shifting algorithms have been widely used for three-dimensional (3D) shape measurement. However, when measuring moving objects, the object motion may distort the linear phase and lead to the phase error. This paper presents an effective phase probability equalization (PPE) method for motion-induced error reduction, which directly equalizes the distorted phase to estimate the linear phase. In general, adequate sampling pixels are needed to compute the accurate phase probability (PP). This paper further presents an improved shifted-phase probability equalization (SPE) method, which combines the original phase and its π-shifted phase to double the sampling pixels. Both simulations and experiments have been conducted to measure objects moving in z-direction, and their results validate that the proposed PPE and SPE methods could effectively compensate the motion-induced error.

3D shape measurement techniques for human body reconstruction

<p>In this work the performances of three different techniques for 3D scanning have been investigated. In particular two commercial tools (smartphone camera and iPad Pro LiDAR) and a structured light scanner (Go!SCAN 50) have been used for the analysis. First of all, two different subjects have been scanned with the three different techniques and the obtained 3D model were analysed in order to evaluate the respective reconstruction accuracy. A case study involving a child was then considered, with the main aim of providing useful information on performances of scanning techniques for clinical applications, where boundary conditions are often challenging (i.e., non-collaborative patient). Finally, a full procedure for the 3D reconstruction of a human shape is proposed, in order to setup a helpful workflow for clinical applications.</p>

High-quality 3D shape measurement with binary half truncated sinusoidal fringe pattern

Digital fringe projection (DFP) with defocused binary fringe patterns could overcome the projector nonlinearity and achieve a high-speed 3D measurement. Most of the binary defocusing techniques take sinusoidal fringe as the target of binarization, hoping the binary fringes can become “quasi-sinusoidal” after defocusing. However, considering that the phase-shifting algorithm (PSA) has an inherent resistance to certain order harmonics components, it is not the best choice to make the defocused binary fringe become more sinusoidal. This paper proposed a new kind of fringe called half truncated sinusoidal (ht-sinusoidal) fringe, which has the same phase result as the sinusoidal fringe with 4-step PSA theoretically. And this kind of fringe is easier to be binarized for the reason that half of it has 0 intensity values already. Therefore it can achieve significant phase quality improvements compared with the binary sinusoidal fringe. We mainly verified that the ht-sinusoidal fringe will help substantially improve the measurement accuracy of different error diffusion methods through both simulations and experiments.

3D shape measurement based on the unequal-period combination of shifting Gray code and dual-frequency phase-shifting fringes

The combination of Gray code and phase-shifting fringes is widely used in the 3D structured light measurement. However, an error in the period order of the phase-shifting fringe obtained by the Gray code results in a period jump error approximately equal to an integer multiple of 2π. This error causes a gross measurement error, and even a measurement failure for a measured surface whose height changes drastically. In order to effectively reduce this error, a 3D measurement method, based on the unequal-period combination of shifting Gray code and dual-frequency phase-shifting fringes, is proposed by analyzing the generation mechanism of the period jump error. In the proposed method, on the basis of the unequal-period combination of shifting Gray code and single-frequency phase-shifting fringe, a strategy of additionally projecting a set of low-frequency phase-shifting fringe patterns is adopted to effectively correct the period jump error. A measurement model of this method is established, its applicable conditions are given, the experimental equipment is built, and verification experiments are carried out. The experimental results show that the proposed method can effectively reduce the period jump error, and can measure multiple isolated objects and objects with drastic changes in surface height such as plaster heads.

High-accuracy point cloud registration for 3D shape measurement based on double constrained intersurface mutual projections

The accuracy of 3D shape measurement directly determines the quality and reliability of intelligent manufacturing, and point cloud registration is the key factor. However, the dislocation of discrete point clouds in the overlapping area seriously reduces the registration accuracy. This paper proposes a new high-accuracy registration method based on double constrained intersurface mutual projections. First, the initial registration set is built by mutual projections between similar local regions, then the final registration set is determined by the rigid transformation consistency constraint, finally, high-accuracy pairwise registration is realized. On this basis, the K-means clustering is further fused to achieve multiview global optimization. On Stanford dataset, both of the pairwise and multiview registration error significantly decreased. In the experiment with GoScanG1 scanner, the surface error of pairwise registration was reduced by 2.2%; that of the multiview registration was reduced by 42.5%, which shows the advancement of the new method in actual measurements.

3D reconstruction from structured-light profilometry with dual-path hybrid network

With the rapid development of high-speed image sensors and optical imaging technology, these have effectively promoted the improvement of non-contact 3D shape measurement. Among them, striped structured-light technology has been widely used because of its high measurement accuracy. Compared with classical methods such as Fourier transform profilometry, many deep neural networks are utilized to restore 3D shape from single-shot structured light. In actual engineering deployments, the number of learnable parameters of convolution neural network (CNN) is huge, especially for high-resolution structured-light patterns. To this end, we proposed a dual-path hybrid network based on UNet, which eliminates the deepest convolution layers to reduce the number of learnable parameters, and a swin transformer path is additionally built on the decoder to improve the global perception of this network. The experimental results show that the learnable parameters of the model are reduced by 60% compared with the UNet, and the measurement accuracy is not degraded at the same time. The proposed dual-path hybrid network provides an effective solution for structured-light 3D reconstruction and its practice in engineering.

A 3D shape measurement method for high-reflective surface based on accurate adaptive fringe projection

Fringe projection profilometry (FPP) is a popular three-dimensional (3D) shape measurement technique with advantages of high-precision and non-contact. However, measuring high-reflective surface is still a challenging task for conventional FPP because image saturation leads to absolute phase errors and reconstruction errors. In this paper, a new adaptive fringe projection (AFP) method is proposed. Compared with existing AFP methods, we simplify the experimental process and get more accurate coordinate mapping. First, we built the projection intensity model for each projector pixel rather than camera pixel. Then, a uniform gray pattern was projected and two images were captured under high and low exposures respectively to calculate the quantitative low projection intensity. In coordinate mapping, sub-pixel coordinates were mapped from camera image onto projector image. They were filtered with absolute phase to deal with resolution difference between camera and projector and get accurate coordinate correspondence. Finally, adaptive fringe patterns were generated. Besides, we proposed a novel quantitative criterion to evaluate the effectiveness of AFP methods called point cloud integrality (PCI), which is calculated based on the number of 3D points of high-reflective areas. The experiments demonstrate that the proposed method increases both the measurement accuracy and PCI compared with conventional FPP and some existing AFP methods.

Advances and Prospects of Vision-Based 3D Shape Measurement Methods

Vision-based three-dimensional (3D) shape measurement techniques have been widely applied over the past decades in numerous applications due to their characteristics of high precision, high efficiency and non-contact. Recently, great advances in computing devices and artificial intelligence have facilitated the development of vision-based measurement technology. This paper mainly focuses on state-of-the-art vision-based methods that can perform 3D shape measurement with high precision and high resolution. Specifically, the basic principles and typical techniques of triangulation-based measurement methods as well as their advantages and limitations are elaborated, and the learning-based techniques used for 3D vision measurement are enumerated. Finally, the advances of, and the prospects for, further improvement of vision-based 3D shape measurement techniques are proposed.

Improved dual-frequency phase-coding fringe projection method for 3D shape measurement

The phase-coding method is useful in 3D shape measurement due to its low sensitivity to surface contrast and camera noise. However, as the number of fringe periods increases, the wrapped phase recovered from the sinusoidal phase-shifting patterns is always misaligned with the fringe order calculated by the stair-coding patterns. To solve this, an improved dual-frequency phase-coding fringe projection method is designed to realize absolute phase retrieval. Specifically, the auxiliary fringe order determined by phase-shifting is used to get the absolute phase, which can eliminate fringe-order errors. Moreover, to increase speed and the number of codewords, this paper presents a scheme that only requires seven patterns, including a uniform gray image, two high-frequency sinusoidal fringe patterns, two low-frequency sinusoidal fringe patterns and two phase-coding fringe patterns. Therefore, this method requires fewer patterns to realize high-resolution measurement than existing methods. The success of the proposed method can be proved by experiments.

Composite fringe projection deep learning profilometry for single-shot absolute 3D shape measurement.

Single-shot fringe projection profilometry (FPP) is essential for retrieving the absolute depth information of the objects in high-speed dynamic scenes. High-precision 3D reconstruction using only one single pattern has become the ultimate goal in FPP. The frequency-multiplexing (FM) method is a promising strategy for realizing single-shot absolute 3D measurement by compounding multi-frequency fringe information for phase unwrapping. In order to solve the problem of serious spectrum aliasing caused by multiplexing schemes that cannot be removed by traditional spectrum analysis algorithms, we apply deep learning to frequency multiplexing composite fringe projection and propose a composite fringe projection deep learning profilometry (CDLP). By combining physical model and data-driven approaches, we demonstrate that the model generated by training an improved deep convolutional neural network can directly perform high-precision and unambiguous phase retrieval on a single-shot spatial frequency multiplexing composite fringe image. Experiments on both static and dynamic scenes demonstrate that our method can retrieve robust and unambiguous phases information while avoiding spectrum aliasing and reconstruct high-quality absolute 3D surfaces of objects only by projecting a single composite fringe image.

Accurate 3-D Shape Measurement for Large Objects Using Speckle-Assisted Fringe Projection and Global Markers Localization

Digital fringe projection (DFP) three-dimensional (3D) shape measurement technique has been extensively researched and applied due to its advantages of high accuracy and high efficiency. To retain the measurement accuracy for large objects, results of multiple measurements are required to be aligned based on the combination of DFP and auxiliary positioning instruments, which decreases the measurement efficiency and limits the practical application. This paper presents an accurate and efficient scheme for large-scale 3D shape measurement based on speckle-assisted fringe projection and global markers localization. Taking advantage of the triple-view geometric constraints and stereo structured-light model, arbitrary 3D surface can be reconstructed with high quality by projecting only four patterns. To align multiple local shape data of large objects without using auxiliary positioning instruments, an accurate data registration method based on global markers reconstruction and localization is presented, which firstly calculates 3D coordinates of all non-coded markers pasted on the object surface by sequential matching, incremental markers reconstruction and global optimization based on bundle adjustment followed by multiple local data registration through local-to-global ICP matching. The experimental results indicate that the proposed method can substantially decrease the accumulated measurement error and achieve commensurate accuracy with the auxiliary positioning methods.

The Absolute Phase Retrieval Based on the Rotation of Phase-Shifting Sequence

The absolute phase retrieval has been the focus of developments in three-dimensional (3D) shape measurement. The traditional absolute phase retrievals based on phase-shifting algorithm mainly utilize extra patterns to generate fringe order map, which decreases the measuring speed and income more phase shift error. In this paper, the cue of fringe order is directly embedded into phase-shifting patterns to minimize the number of projecting patterns. Fringe order is encoded by rotating the sequence of phase shift amounts, causing the 2&#x03C0; discontinuity to be extended into &#x00B1;2&#x03C0;/<i>N</i>, &#x00B1;4&#x03C0;/<i>N</i>,..., &#x00B1;2(<i>N</i>-1)&#x03C0;/<i>N</i>, &#x00B1;2&#x03C0;. Due to the coding periodicity, the regional wrapped phase without 2&#x03C0; ambiguity can be extracted from different wrapped phase maps. When decoding, the coding phase jump points can be located and identified from one or multiple wrapped phase maps, and then used to determine fringe order by the bidirectional data processing. Experimental results have proved that the proposed method without extra coding patterns not only provides high robustness to shadow and considerable accuracy closed to that of traditional phase-coded method, but also inherits the performance to measure the complex and colorful object using traditional sinusoidal phase-shifting patterns.

Deep-learning-enabled dual-frequency composite fringe projection profilometry for single-shot absolute 3D shape measurement

Single-shot high-speed 3D imaging is important for reconstructions of dynamic objects. For fringe projection profilometry (FPP), however, it is still challenging to recover accurate 3D shapes of isolated objects by a single fringe image. In this paper, we demonstrate that the deep neural networks can be trained to directly recover the absolute phase from a unique fringe image that involves spatially multiplexed fringe patterns of different frequencies. The extracted phase is free from spectrum-aliasing problem which is hard to avoid for traditional spatial-multiplexing methods. Experiments on both static and dynamic scenes show that the proposed approach is robust to object motion and can obtain high-quality 3D reconstructions of isolated objects within a single fringe image.

An Optimized Dithering Approach for 3d Shape Measurement Using Phase-Based Chaos Genetic Algorithm

An Optimized Dithering Approach for 3d Shape Measurement Using Phase-Based Chaos Genetic Algorithm

High-speed three-dimensional shape measurement with inner shifting-phase fringe projection profilometry

Fringe projection profilometry (FPP) has been extensively studied in the field of three-dimensional (3D) measurement. Although FPP always uses high-frequency fringes to ensure high measurement accuracy, too many patterns are projected to unwrap the phase, which affects the speed of 3D reconstruction. We propose a high-speed 3D shape measurement method using only three high-frequency inner shifting-phase patterns (70 periods), which satisfies both high precision and high measuring speed requirements. Besides, our proposed method obtains the wrapped phase and the fringe order simultaneously without any other information and constraints. The proposed method has successfully reconstructed moving objects with high speed at the camera’s full frame rate (1700 frames per second).

Self-Unwrapping Phase-Shifting for Fast and Accurate 3-D Shape Measurement

Phase unwrapping is one of the most important procedures in three-dimensional phase measurement profilometry and has attracted great research interest in the past few decades. However, existing phase unwrapping methods require additional patterns or pre-assumptions to determine the fringe orders for further absolute phase retrieval, which dramatically affect the measurement efficiency. To address this problem, we propose a generic self-unwrapping phase-shifting algorithm that retrieves the absolute phase without external information or priors. To this end, we firstly embed a novel space-varying phase shift that uniquely determines the fringe order information into sinusoidal patterns, and then extract it to retrieve the absolute phase by pixel-wise calculation. Our method achieves higher efficiency over previous methods while preserving the measurement precision, and the minimum number of patterns required is only four. All theoretical, simulation and extensive experimental results demonstrate its superiority in fast and accurate 3D shape measurement.

Omnidirectional 3d Shape Measurement Using Image Outlines Reconstructed from Gabor Digital Holography

Omnidirectional 3d Shape Measurement Using Image Outlines Reconstructed from Gabor Digital Holography