Hongzhi Jiang

High-speed triangular pattern phase-shifting 3D measurement based on the motion blur method

Recent advancements in 3D measurement technologies have increased the urgency of requiring high-speed 3D measurement in many fields. This study presents a novel four-step triangular pattern phase-shifting 3D measurement using the motion blur method, which combines the advantages of phase-shifting methods. To comply with the high speed requirement, binary coded triangular patterns are projected and could dither vertically. Therefore, the image captured by the camera is blurred into grayscale-intensity triangular patterns, which can be used for phase unwrapping and 3D reconstruction. The proposed method decreased the projection time compared with sinusoidal patterns using a DMD (digital micromirror device) projector. Furthermore, this study presents a four-step triangular phase-shifting unwrapping algorithm. The experiments indicate that the proposed method can achieve high-speed 3D measurement and reconstruction.

Automatic evaluation of machining allowance of precision castings based on plane features from 3D point cloud

A novel automatic precision casting machining allowance evaluation approach, which is accomplished by a two-step rough-precise point cloud registration based on plane features extracted from the two point clouds (i.e. the measured precision casting point cloud and the point cloud discretized from the CAD model), is proposed in this paper. Firstly, the two point clouds are registered roughly by PCA algorithm. Secondly, an improved plane fitting and merging algorithm is proposed to extract the plane features from both the two point clouds. The extracted plane features are matched by searching the nearest plane feature description vector. The rotation matrix for the precise registration can then be derived by registering the normal vectors of the matched plane features. Finally, the machining allowance at each point is obtained by calculating the distance between the corresponding points along the normal direction. The experiment on precision casting machining allowance evaluation is given to show the performance of the proposed approach.

Adaptive regional single-pixel imaging based on the Fourier slice theorem

Single-pixel imaging (SPI) is a novel method for capturing high-quality 2D images of scenes using a non-spatially-resolved detector. While implementing conventional SPI, a huge number of illuminated patterns are projected onto the object to reconstruct a sharp image. For a situation in which the object occupies part of the illuminated region, we propose an adaptive regional SPI method (ARSI) to decrease the number of projected patterns. In the ARSI scheme, the object region is adaptively located based on the Fourier slice theorem. Then, the illuminated patterns are projected only onto the object region to facilitate imaging efficiency. Experiments demonstrate that the proposed ARSI method can achieve sharp image reconstruction with a substantial reduction in pattern number, thereby improving imaging efficiency.

Hyper thin 3D edge measurement of honeycomb core structures based on the triangular camera-projector layout & phase-based stereo matching

We propose a novel hyper thin 3D edge measurement technique to measure the profile of 3D outer envelope of honeycomb core structures. The width of the edges of the honeycomb core is less than 0.1 mm. We introduce a triangular layout design consisting of two cameras and one projector to measure hyper thin 3D edges and eliminate data interference from the walls. A phase-shifting algorithm and the multi-frequency heterodyne phase-unwrapping principle are applied for phase retrievals on edges. A new stereo matching method based on phase mapping and epipolar constraint is presented to solve correspondence searching on the edges and remove false matches resulting in 3D outliers. Experimental results demonstrate the effectiveness of the proposed method for measuring the 3D profile of honeycomb core structures.

A High Throughput Integrated Hyperspectral Imaging and 3D Measurement System

Hyperspectral and three-dimensional measurements can obtain the intrinsic physicochemical properties and external geometrical characteristics of objects, respectively. The combination of these two kinds of data can provide new insights into objects, which has gained attention in the fields of agricultural management, plant phenotyping, cultural heritage conservation, and food production. Currently, a variety of sensors are integrated into a system to collect spectral and morphological information in agriculture. However, previous experiments were usually performed with several commercial devices on a single platform. Inadequate registration and synchronization among instruments often resulted in mismatch between spectral and 3D information of the same target. In addition, using slit-based spectrometers and point-based 3D sensors extends the working hours in farms due to the narrow field of view (FOV). Therefore, we propose a high throughput prototype that combines stereo vision and grating dispersion to simultaneously acquire hyperspectral and 3D information. Furthermore, fiber-reformatting imaging spectrometry (FRIS) is adopted to acquire the hyperspectral images. Test experiments are conducted for the verification of the system accuracy, and vegetation measurements are carried out to demonstrate its feasibility. The proposed system is an improvement in multiple data acquisition and has the potential to improve plant phenotyping.

Defocusing parameter selection strategies based on PSF measurement for square-binary defocusing fringe projection profilometry

Three-dimensional (3D) shape measurement system with binary defocusing technique can perform high-speed and flexible measurements if binary fringe patterns are defocused by projector properly. However, the actual defocusing degree is difficult to set, and the fringe period is difficult to determine accordingly. In this study, we present a square-binary defocusing parameter selection framework. First, we analyze the fringe formation process mathematically. The defocusing degree is quantified and manipulated by using the focusing distance of projector, which is calibrated by point spread function measurement. To optimize parameter selection, single-point sinusoidal error is modeled as the objective function for the evaluation of the defocusing effect. We verify the correctness by using different parameter combinations and object measurements in our experiments. The appropriate defocusing parameters can be easily obtained according to the analysis of practical system setup, which improves the quality and robustness of the system.

Projector-defocusing rectification for Fourier single-pixel imaging

Fourier single-pixel imaging (FSI) is an efficient single-pixel imaging method of obtaining high-quality (resolution/signal-to-noise ratio) 2D images, which projects sinusoid patterns on the object and reconstructs the image through reflected light. The typical system of FSI consists of a single-pixel detector and a digital projector. However, the defocusing of the projector lens blurs the projected patterns, which results in reduced imaging quality. In this work, we propose the projector-defocusing rectification for FSI, which optimizes projector defocusing for the first time. The proposed method rectifies Fourier coefficients using the amplitude ratio between original and defocused patterns, which we can acquire through a controlled experiment on a whiteboard. The enhancement of imaging quality in imperfect circumstances is demonstrated by simulations and experiments.

A dual-direction fringe projection method for the 3D measurement of translucent object

The fringe projection technology is widely used in 3D measurement fields. However when the technology is applying to translucent objects, the subsurface scattering and absorbing always leads to a decline of the measurement accuracy. The aim of this paper is to propose a dual-direction fringe projection method in order to obtain an more accurate measurement result for the translucent objects as while as change the whole measurement system little and do not reduce the measuring rapidity. The paper mainly includes three parts: (1) The principle of dual-direction fringe projection method and different forms of dual-direction fringe; (2) Analysis of the different effect for the measurement accuracy brought by different factors; (3) Experiments for artificial tooth by various dual-direction fringes and accuracy analysis. The experiment results showed that by this method it is possible to improve the measurement accuracy for the translucent objects.

The 3D measurement techniques for ancient architecture and historical relics

Nowadays, 3D measurement and re-construction technologies are widely used not only in industry area, but also in the appreciation and research of ancient architecture and historical relics. Many methods are used for the architecture measurement in large scale, but as for the details of architecture or precision historical relics, these methods meet difficulties. Thus, historical relic objects with specular surface or complex sculptural surface could not be measured by traditional method. Focusing on these problems, this paper proposed 3D measurement technique which contains two levels of measurement. Firstly, when measuring ancient architecture in large scale, laser scanning and photometry methods are used. Then, when measuring details of architecture, a fast and adaptive 3D measurement system is used. Multi-view registration is also used for the measurement of hollowed-out structure of sculptural relics. The experiments indicate that the system can achieve 3D measurement and re-construction of different types of ancient architecture and historical relics.

Calibration of AOTF-based 3D measurement system using multiplane model based on phase fringe and BP neural network

A specifically designed imaging system based on an acousto-optic tunable filter (AOTF) can integrate hyperspectral imaging and 3D reconstruction. As a result of the complicated optical structure, the AOTF imaging system deviates from the traditional pinhole model and lens distortion form, causing difficulty to achieve precise camera calibration. The influencing factors leading to the deviation are discussed and a multiplane model (MPM) is proposed with phase fringe to produce dense mark points and a back propagation neural network to obtain subpixel calibration. Experiments show that MPM can reduce the back projection error efficiently compared with the pinhole model. A 3D reconstruction process is conducted based on the calibration result to verify the feasibility of the proposed method.