This PDF file contains the front matter associated with SPIE Proceedings Volume 11571, including the Title Page, Copyright information, and Table of Contents.

The status of ocean remote sensing is becoming more and more important. There are all kinds of resources in ocean and many ships on the sea. It is necessary for people to detect or observe these objects to know more about the natural resources and to ensure safety of ships. The paper concentrated on a compact design of a small space camera. An oscillating mirror was designed in the camera for a small volume. After that, structure analysis was done by traditional method and reduced method. The key step of reduced method was to get the reduced model. The analysis result showed that the reduced method could not only cut down the cost of computing, but also give a result with good accuracy. Besides, the analysis result indicated that the small space camera could undergo the strict load cases smoothly. The design in this paper may give some guidance to other designers and engineers when they are going to make a small space camera for ocean remote sensing.

Over the past decades, numerous efforts have been made to improve imaging resolution beyond the Abbe-Rayleigh limit. Optical super-oscillations, the phenomenon that the certain combination of band-limited functions could oscillate arbitrarily faster than its highest Fourier components, offer a promising route to optical super-resolution imaging. Here we developed a super-oscillation function filter based on super directivity array antenna, which is used in digital image processing. We demonstrate the ability of this method in recovering coherent diffraction limit images. It was proved that two holes separated by a distance about 0.7 times of the diffraction limit can be distinguished, and it is consistent with the capacity of the filter designed.

Imaging blur is an inevitable problem because of the low response to medium frequency in optical multi-aperture imaging system, in which Wiener filtering is usually used in implementing image restoration to obtain clear highresolution images. Recent notable developments in the field of deep learning have opened up exciting avenues inspiring us to use data-driven approach for image deblurring in optical multi-aperture imaging system. In this paper, a deep learning framework named RestoreNet, which is based on a U-shaped convolution neural network, is proposed to replace the general Wiener filtering for image restoration. Numerical simulation and experiment results show that RestoreNet could recover the imaging map from system successfully, just like Wiener filtering does. However, RestoreNet only needs one dataset containing a few images for training, and shows strong image restoration ability without the point spread function or optical transfer function of system in testing, as well as the priori information of object and noise. As a result, RestoreNet is an effective alternative in image restoration of the optical multi-aperture imaging system.

We using a problem in mechanics, such as fastest sending messenger to introduce the concept of equivalent distance. We analyzed the choice of a path in kinematics problem. Combining the application of Fermat's principle in the law of refraction, we established the corresponding analogy relationships of optical and kinematics physical description. The analogy method presented in this paper would be helpful for deep understanding of optical concepts such as: optical path etc. And provide a deeply understanding reference method for Fermat’s theorem. especially in a complex stratified media.

Aiming at the problem of mutual restriction between the quality of image and the equipment size in the current near-eye display (NED) system, we propose a NED system based on metasurface. We design a colorful NED system for MR/AR applications using all-dielectric metasurface and retina display technology. Based on the principle that metasurface can accurately control the wavefront, the system presents high-quality images while achieving lightweight and compact configuration like an ordinary pair of glasses. The semi-transparent metasurface with a size of 10mm×10mm is placed 15mm away from the eyes. Results show that the system could effectively enable monochromic and colorful object display onto the retina, and therefore allow clear images to be seen. With further optimization design, such system could provide a feasible solution for colorful NED system with miniaturization.

Aiming at the problem that the Geiger-mode Avalanche Photodiode (GM-APD) is susceptible to background noise and the detection effect decreases during the day, based on the polarization GM-APD detection model, a set of GM-APDbased polarized lidar imaging experimental equipment is proposed. Using this equipment to image the target in the simulated sunlight environment, the effect of polarization detection on the echo triggering performance was studied. The results show that, compared with the non-polarized system, the polarized system reduces the impact of noise on target detection, improves the image quality, and reduces the false alarm probability of the image. When the laser single pulse energy is 400nJ, and the polarization angle is 135° , the trigger probability of the metal plate is increased by 10.5%, and its false alarm probability is reduced by 4.6%; the trigger probability of the rough wood board is reduced by 15.2%, and its false alarm probability is reduced 8.9%.Due to the deterioration of imaging results caused by background noise, it is proposed to use polarization degree images for further background filtering to extract a more complete target contour. This research work provides experimental support for the effective detection of GM-APD lidar during the day.

In this work, based on the principle of blind deconvolution, and considering the inherent structure of the blur kernel, an automatic relevance determination (ARD) model is used to determine the prior model based on the gradient, instead of intensities of the blur kernel. The results show that the proposed ARD model on gradient can improve the kernel quality and thus produce better de-blurred image. Compared with the case of using the intensity for ARD, the proposed algorithm gives better blur kernel estimation and show robustness against non-Gaussian noise. As concrete examples, we demonstrate that the proposed method is applicable to image restoration on the scenario of camera shake, object motion and defocused blurring.

Single-pixel camera，taking advantage of the technique of compressive sensing, is a new imaging scheme proposed in recent years. It essentially uses a single-point detector to replace the traditional array detector of the camera and a few measurements to complete the image reconstruction. In this paper, we propose a dual-channel acquisition system based on the bidirectional reflection characteristics of spatial light modulator and build such a compressive imaging configuration with substantial noise suppression. In addition, we conducted a video reconstruction study based on this system and compared two reconstruction strategies. One is based on video frame decomposition, which decomposes the video sequence into individual video frames and converts the video reconstruction into a sequence of static image reconstructions. The other is to design a fusion matrix, using its low-dimensional matrix for video preview to obtain motion estimation, and then obtaining the reconstructed video through its high-dimensional matrix. We compared the performance of the two approaches by analyzing the data from the actual experimental platform, and realized real-time video capturing and recovery at 40 fps at resolution 128*128.

The complex amplitudes are usually extracted by windowing the Fourier spectra in the off-axis digital holography. However, it often needs manual operation. In order to obtain the complex amplitude automatically from a single off-axis hologram, the spatial carrier frequency phase shifting method is used in this work. Based on the fact that the complex amplitude to retrieve usually vary slowly and the carrier frequency keeps constant, three consecutive phase-shifting interferograms are obtained using the neighboring pixels along the same line in the hologram. Then the complex amplitude is calculated out. Without Fourier transforms, the computational complexity here is much smaller than that of the traditional one. Simulative and experimental results show that this method can really improve the processing efficiency during the complex amplitude reconstruction, and it can be used for automatic application of the digital offaxis holography.

Adaptive fringe-pattern projection (AFPP) method is an improvement of fringe projection profilometry (FPP) technique. AFPP method avoids image saturation via reducing the intensity of projection pixels which are corresponding with the shiny areas of the object. To enhance the efficiency of the method, in this paper, we proposed an improved AFPP method that reduces the number of patterns to be projected in preliminary steps. In our proposed method, we introduce epipolar constraints when we match the corresponding pixels between the projector and camera images. We only need to project one direction of phase-shifting fringe patterns instead of two directions, so that the time consumption is reduced by at least half in preliminary steps. The experiment result shows that the proposed matching method guarantees the matching accuracy, and the measurement results are more complete via the improved AFPP method.

Quantitative phase imaging technology has become an important technology for imaging phase objects such as cells in virtue of the advantages of non-destructive, non-invasive and high accuracy. More importantly, the phase of a cell is closely related to its internal structure and component. In this paper, according to the morphological and optical characteristics of typical leukocytes, the approximate models of different kinds of cells were firstly established to obtain the corresponding phases. Then, digital image edge operators such as gradient, gradient module and Laplace were used to detect the phase maps. Subsequently with the phase detection images of the cell, its feature information, such as the morphology and the refractive index change can be analyzed. Simulation results suggested that the edge operators are sensitive to the morphology, the interface of each medium and component of the cell. Therefore, the edge detection method of phase image can be used for cell’s classification recognition and preliminary determination of internal structure

A new orbital angular momentum (OAM) model detection approach is proposed by extending the coherent diffraction ptychographic imaging (CDPI) approach. A beam carrying an OAM is a vortex beam which has a doughnut-shape intensity and a helical phase with a topological charge same as the mode index of OAM. To recover its phase helix, a vortex beam radiates a template region-by-region, in grid fashion or scanning way, and the adjacent radiated regions should be overlapped partially. The exit wave field of each radiated region produces a far-field diffraction intensity pattern at the back focal plane of a lens. After recording the intensity patterns, the wavefront of the incident beam can be reconstructed from the patterns via ptychographic iterative engine (PIE). In this paper, we first determine the diffraction intensity recording requirements by analyzing diffraction features of a vortex beam. We second derive the spatial sample requirements in ptychographic OAM detection, for a convergence to the non-ambiguity solution of a helical wavefront, involving two case, with and without a prior knowledge about the template function. Finally, numerical and optical experiments confirm that the helical wavefront of a vortex beam can be accurately recovered by conventional PIE when template is treated as reference. And using a modified regularized PIE (rPIE), the wavefront of incident vortex beam and the unknown transmission function of template can be reconstructed simultaneously.

Phase extraction is a key step in phase imaging. The accuracy and speed of phase recovery directly affect the quality and efficiency of phase imaging. In this paper, a statistical average method for phase retrieval is proposed by combining the two-step phase-shifting interference with the characteristics of diffractive phase field. The main operations of this method are subtraction and statistical average. The simulation results of a sphere show the feasibility and high accuracy of the method. And the simulation results of a red blood cell and a white blood cell discuss the application of this method in biological imaging.

With the widespread use of AR/VR head-mounted displays, there is an increasing demand for pinhole cameras to test their image quality. However, it has been a challenge to design a pinhole camera with a wide field of view (FOV), high image performance and low distortion while maintaining compact and lightweight. In this paper, a pinhole camera optical system is designed. Based on a 1 in. sensor, the pinhole lens has a FOV of 100° and an entrance pupil diameter of 4mm. In our final design, the distortion is limited within 2.5% and the modulation transfer function values across the entire field are above 0.3 at 161 line pairs/mm (lps/mm). The spherical aberration, color aberration and field curvature are corrected. The design result shows that the optical system of the pinhole camera is compact and meets the requirements of the VR/AR image quality test system.

Recently, non-line-of-sight (NLOS) imaging is a novel optical computational imaging technology developed. In NLOS imaging, objects are reconstructed by analyzing information carried in the reflected light. Considering the flaws in NLOS such as information loss and weak target detection due to diffuse reflection of light, developing a stable and effective imaging system is nontrivial. Therefore, in this paper, a 2D object inspection system based on compressed sensing and Orthogonal Matching Pursuit (CSOMP) approaches for effective NLOS imaging is proposed. This will not only alleviate the difficulty in collecting object information in a low light environment but also make the reconstruction process with a few measurements. Specifically, the single-pixel camera based on Digital Micromirror Device (DMD) is introduced to fully exploit the programmable pattern information and the correlation between the measured intensity data for image renovation. To reduce acquisition time, a high-efficiency single-pixel detector is employed that obviates the need for mechanical scanning. Also, in the image reconstruction stage, the residuals of the measurement matrix are obtained through OMP. The proposed method combines with the CS theory; thus, the target can be rebuilt with fewer measurements. Experimental results on NLOS images illustrate that the proposed imaging scheme reconstructed the image exactly similar to the real object.

The traditional reflective parallel optical tube has a small field of view and a large off-axis aberration, so it is difficult to meet the high precision requirements of wide field of view and wide band. The design adopts the initial structure of wide-field eyepiece, and is optimized by the optical design software.The focal length of the system is 307mm, the field of view Angle is ±12°, the full field of view has no halo, the working wavelength is 450nm~900nm, and the transfer function is almost close to the diffraction limit, meeting the design requirements of large field of view, wide band and small distortion.

High resolution and accuracy phase modulation could help to improve the diffraction efficiency of spatial light modulator. For spatial light modulator with analog driving pixels, the phase accuracy is mainly decided by the DAC design. The conventional circuit design method to improve the DAC accuracy is using a large chip area or using expensive advanced manufacturing technology. In this paper, a calibration method with both analog and digital algorithm is presented. By adding digital calibration and improving the frame frequency, the liquid crystal response characteristic is improved without changing the on chip DAC design. An 8 bit DAC LCoS panel is used for verification. The experiment result shows, the phase error is reduced from 1.49% to 0.44%, and the diffraction efficiency is improved from 72% to 75%. And a holography picture is projected to verify the effect.

Coherent diffractive imaging (CDI) is a lensless, high-resolution imaging method. In order to overcome the problems of poor-quality image and slow convergence of iterative algorithm in traditional CDI, here multimode fiber source is selected as the lighting source to replace the traditional plane-wave. A phase-only spatial light modulator (SLM) is used as a negative lens instead of a positive lens in a 4f imaging unit. The diffraction field can be rapidly changed at a very short distance so as to realize a fast convergence and then save a lot of iteration times. This proposed setup has advantages such as a simpler structure and static recording. Both simulations and visible-light experiments demonstrate the good imaging quality and robustness of this method.

Volume holograms can record and reproduce three-dimensional object. Compared with planar holograms, volume holograms can improve the resolution, stereo perception and realism of anti-counterfeiting, and realize three-dimensional anti-counterfeiting. In order to improve the diffraction efficiency and anti-counterfeiting effect of volume holographic anti-counterfeiting, this paper proposes a reflection volume holographic three-dimensional anti-counterfeiting method based on photopolymer. The diffraction characteristics and influencing factors are analyzed by Kogelnik’s coupled wave theory. Based on Piazzolla’s monomer diffusion model, the effects of exposure energy and exposure intensity on refractive index modulation and diffraction efficiency were studied. The simulation results show that the reflection volume holograms can achieve a greater refractive index modulation, with higher diffraction efficiency, stronger wavelength selectivity, which enhances the effect of three-dimensional anti-counterfeiting.

Volume holograms can achieve concentration of sunlight, but it has shortcomings such as small concentration angle, dispersion, and complex production process of the volume holography material, which is difficult to mass produce. This paper proposes a method based on the metasurface concentrating sunlight. We analyzed the mechanism of volume holographic light focusing by Kogelnik coupling wave theory, which is greatly limited by wavelength selectivity. Utilizing the metasurface control mechanism on the wave front phase, the array arrangement that meets the light concentrating effect was designed. Simulation analysis shows that the metasurface can achieve light concentration in the visible light band with low dispersion.

In this work, we propose a learning-based absolute 3D shape measurement based on single fringe phase retrieval and speckle correlation. Our method combines the advantages of Fourier transform profilometry (FTP) techniques for high-resolution phase retrieval and speckle correlation approaches for robust unambiguous depth measurement. The proposed deep learning framework comprises two paths: one is a U-net-structured network, which is used to extract the wrapped phase maps from a single fringe pattern with high accuracy (but with depth ambiguities). The other stereo matching network produces the initial absolute (but with low resolution) disparity map from an additional speckle pattern. The initial disparity map is refined by exploiting the wrapped phase maps as an additional constraint, and finally, a high-accuracy high-resolution disparity map for absolute 3D measurement can be obtained. Experimental results demonstrated that the proposed deep-learning-based method could realize high-precision absolute 3D measurement for measuring objects with complex surfaces.

Absolute phase unwrapping is important for optical phase-based three-dimensional (3D) measurements of isolated objects or discontinuous surfaces. For fringe projection profilometry, the fringe order of each stripe can be obtained uniquely by temporal phase unwrapping (TPU) that employs at least two different phase maps of the same measured scene. In this work, we present a novel TPU strategy using deep learning. Our idea is to treat the calculation of fringe order as a classification problem, which can be solved by a lightweight fully connected neural network. Consequently, the training process of our network can be finished within an hour, which saves a large amount of time and makes it possible to deploy the network on mobile devices. Moreover, rather than obtaining the training data from time-consuming real experiments, we simulate the data with software under different noises.

X-ray Ptychography is a high-resolution coherent diffraction imaging method, which has been widely used in the third generation of synchrotron radiation sources. In this method, diffraction images are collected with a series of probe incident positions on an extended sample with partial overlapping between neighbored exposure areas. Accurate location information of the probe is very crucial to obtain high-resolution reconstruction images in experiment. However, the motor movement errors are usually unavoidable, which are caused by thermal drift or mechanical inaccuracy. We proposed an effective method to correct the scanning position errors, which combined the Fresnel CDI algorithm and the image registration method. The feasibility and robustness of the algorithm have been verified by both simulation and xray experiments.

We propose label-free and motion-free resolution-enhanced intensity diffraction tomography (reIDT) recovering the 3D complex refractive index distribution of an object. By combining an annular illumination strategy with a high numerical aperture (NA) condenser, we achieve near diffraction-limited lateral resolution of 346 nm and axial resolution of 1.2 μm over 130 × 130 × 8 μm³ volume. Our annular pattern matches the system’s maximum NA to reduce the data requirement to 48 intensity frames. The reIDT system is directly built on a standard commercial microscope with a simple LED array source and condenser lens adds-on, and promises broad applications for natural biological imaging with minimal hardware modifications. To test the capabilities of our technique, we present the 3D complex refractive index reconstructions on an absorptive USAF resolution target. Our work provides an important step in intensity-based diffraction tomography towards high resolution imaging applications.

In the long-distance imaging system, one of the main factors limiting the imaging resolution is the size of the imaging lens aperture, which determines the diffraction limit of the optical system. Therefore, we propose a non-interference synthetic aperture super-resolution imaging reconstruction and optimization method. The camera array is used to collect a series of low-resolution sub-aperture images. Combined with Fourier ptychography imaging algorithm, the spectrum and aperture function of the current sub-aperture diameter is updated by using the optimization algorithm based on adaptive step size. to obtain the high-resolution spectrum information of the target to be measured. Meanwhile, the high-resolution spectrum information of the target is obtained. In the reconstruction process, the simulated annealing algorithm is introduced to correct the positioning error of the sub-aperture, and the optimization algorithm is used to update the sub-aperture, which greatly improves the accuracy of the reconstruction results and achieves the theoretical imaging resolution. Moreover, it also has excellent imaging results for complex objects, which verifies the feasibility of the algorithm.

Recently, with the increasing demand for high-accuracy three-dimensional (3D) imaging and localizing of the internal subcellular structures in cells, quantitative refractive index tomography technique has been developed and widely applied in various fields, such as molecular biology, biochemistry and cell biology. Thanks to the noninterference, large field of view and high-resolution advantages of traditional Fourier ptychographic microscopy (FPM) technique, quantitative phase tomography (QPT) method based on multi-slice (MS) FPM has attracted considerable attention lately. However, since the reported MS model is simply established on the traditional two-dimensional FPM, it is difficult to characterize the 3D diffractive propagation properties under large oblique illumination angles accurately. In order to solve this problem, an improved oblique phase correction function for 3D MSFPM is proposed in this paper. The numerical phase error under the large oblique illumination angle is compensated in the iterative reconstruction algorithm, and then the reconstruction accuracy of the 3D QPT can be improved.

Phase space is an artificially structured space jointly represented by the space and spatial frequency variables, and the Wigner distribution function (WDF) is widely used to describe the phase space. Compared with the coherent optical field expressed by two-dimensional (2D) complex amplitude, the four-dimensional WDF (2D space and 2D spatial frequency) can characterize optical field with arbitrary coherent state (coherent, partially coherent and incoherent). It is especially advantageous for the characterization of partially coherent optical fields. The WDF is real and may have negative values, which are the result of phase-space interference. The direct measurement of phase space can only obtain the absolute value of the whole WDF because the camera sensors can only obtain intensity, therefore, the negative values of measured WDF are mixed with positive values. In this paper, an improved phase-space retrieval method is demonstrated. First, capture three-dimensional intensity focal stack by camera sensors. Then, phase space tomography (PST) combined with a non-linear iterative algorithm is conducted to reconstruct the whole WDF.

Most of the traditional microscopic systems are based on the structural design of a single telecentric camera combined with a single projector. The projector in such system needs to be accurately calibrated as an additional view to complete the 3D measurement with the camera. However, it is difficult to achieve high-accuracy calibration due to problems such as gamma, lens distortion, and additional lenses that are difficult to accurately align. In this work, we present a structural design of microscopic 3D measurement system based on dual-telecentric optical path, which contains two camera that can be accurately calibrated and a projection light path. Our design can make full use of the advantages of the full-field structured light projection and is convenient to calibrate. Two different perspectives can provide 3D data from different depth and multiple angles simultaneously so that measurement volume can be enlarged by superimposing the data.

Large viewing angle, dense viewpoints, high-resolution 3D light field display technology can achieve large viewing angle, high resolution, and high dynamic reconstruction of the objective world scene light field, which is an important direction for the development of future display technology. The application of advanced 3D light field display technology is getting more and more attention. However, the current lack of efficient access to large viewing angles, dense viewpoints, and high-resolution 3D light field content limits its development and application. To this end, here proposes a dense viewpoints video generation, coding, transmission method, and build a super multi-view 3D light field video live broadcast system. The method and system are composed of 3D light field video generation, light field video coding, light field video transmission and a display part. It can provide viewers with a real-time, high-quality 3D visual experience.

Fringe projection profilometry (FPP) is one of the most important optical non-contact three-dimensional (3D) measurement technologies. However, in order to satisfy human beings’ visual perception, gamma is artificially added to the digital projector. In past decades, researchers have made efforts to compensate gamma nonlinear errors, but how to efficiently and conveniently correct the gamma distortion is still a big challenge. Inspired by the successful application of deep learning in FPP, we propose a deep-learning-based gamma compensation method. Through extensive data set training, the neural network can learn to acquire the distortion-free high-quality phase information from the phase-shifting images with gamma. Experimental results demonstrate that our method can effectively compensate gamma-induced phase errors, and thus improve the measurement accuracy.

Fourier Ptychographic Microscopy (FPM) has achieved large field of view and high-resolution microscopy, which has attracted widespread attention. However, whether FPM can be restored accurately has not been able to provide a theoretical guarantee. The corresponding measurement system of FPM is the phase retrieval problem of the 2-dimensional inverse short-time Fourier transform (2-D ISTFT). In this paper, aiming at the defect that there are multiple global optimal solutions (ambiguities) for the phase retrieval problem, it is proved that FPM has almost the unique global optimal solution in the sense of removing the global phase solution. Based on different overlap ratios (the ratio of the area of the overlapped part to the window area), the upper bound estimation of the number of the ambiguities is given, and it is proved that FPM can eliminate the conjugate flip solution. In the simulations, the sequential Gerchberg-Saxton method is used to update under different overlap ratio conditions. It is verified that under the condition of low overlap ratio, FPM has poor convergence; under the condition of high overlap ratio, FPM reaches the global optimal solution rapidly.

Using terahertz time-domain spectroscopy (THz-TDS) to characterize the shale is significant for geological research. This paper adopts the transmission THz-TDS system which is composed of a femtosecond laser, terahertz radiation generation device, corresponding detection device, and time delay control system. It can also further explore the application potential of THz-TDS in rock mineralogy. The method of shale powder tablet is used to make sample suitable for test and confirm the appropriate proportion that the shale mixed with PTFE under the condition of THz-TDS. Matlab software is used to extract the phase shift and amplitude curve of the sample, and then the refractive index and absorption coefficient of the sample are extracted from the time domain data. In addition, it is feasible to use the characteristic peak in the absorption coefficient to identify minerals in similar shales. The mineral composition of shales which cannot be directly determined by THz-TDS is determined by X-ray fluorescence (XRF) and X-ray diffraction (XRD). The experimental results show that the refractive index can be used to distinguish two kinds of similar black shales: oil shale which is rich in organic matter and ordinary shale. THz-TDS can be used as a supplement to traditional testing methods, which provide a new method for the study of the rock mineralogical characteristics of shale and make the characterization system of shale more perfect.

Ground based optical observation is widely used in the field of astrometry of moving objects such as Near-Earth Objects and space debris due to its low cost and easy deployment. However, it’s difficult for traditional optical telescopes to have both large aperture and wide FOV, making it impossible to achieve both observation depth and observation efficiency. We propose the computational imaging astrometry method to solve this problem. Based on a scientific CMOS detector and a small-aperture telescope with wide FOV, through strict correction of various astronomical effects and instrumental errors, we establish a high-precision overlay enhancement method for wide-FOV astronomic images. Including displacement overlay of high frame rate images at sub-pixel scale. In the case of no mechanical tracking device, the computational imaging astrometry method enables the observation system to simultaneously track multiple types of moving objects, and exert the ability of image overlay to reduce background noise, thus improving the SNR of the target and enhancing the observation depth of the system.

In this paper, a trinocular vision measurement method of visible cameras with wide field of view is proposed for the limitations of binocular stereo vision technology, such as blind spot in some areas with large measuring angles, mismatching of feature points location and limited field of view. Firstly, three visible light cameras arranged in an equilateral triangle arrangement are used to obtain images with a field of view of 60°×60°. Then, the measurement method based on trinocular stereo vision is researched: FAST algorithm is used to detect the feature points of the three images and the method of tri-crossing polar line constraint is used to match the feature points. Then the RANSAC method is used to calculate the position and attitude with the world coordinate system. After all, the proposed method is demonstrated by some experimental results. This method provides a solution for the relative pose measurement of noncooperative target in complex scenes/sparse features, and the problem of high matching rate based on image similarity matching in binocular stereo vision measurement can be avoided by using the tri-crossing polar line constraint, thus higher pose resolution accuracy can be obtained.

Recent researches on object segmentation mostly concentrate on single-view images or objects in 3D settings. In this paper, a novel method for efficient multi-view foreground object segmentation is presented, using spatial consistency across adjacent views as constraints to generate identical masks. Even though the conventional segmentation results at different views are relatively accurate, there always are inconsistent regions where the boundaries of the mask are different over the same area across multiple views. The central idea of our method is to utilize the camera parameters to guide the refocusing procedure, during which each instance across different views is refocused using multi-view projections. The refocused images are then used as the input of instance segmentation network to predict the bounding box and object mask. The final step takes the network output as the prior information for the GMMs to achieve more accurate segmentation results. While many concrete implementations of the general idea are feasible, satisfactory results can be achieved with this simple and efficient approach. Experimental results demonstrate both qualitatively and quantitatively that the proposed method outputs excellent results with less background pixels, thus allowing us to improve the 3D display quality eventually. We hope this simple and effective method can be of help to future researches in relevant tasks.

In this paper, a holographic lensless quantitative phase imaging (QPI) microscope is presented, which is composed of a CMOS detector image sensor with a programmable color LED matrix, without any lens and mechanical displacement device. Such a miniaturized system can provide a field-portable cost-effective platform for highthroughput quantification of multiple samples. Coordinating the self-developed software operating system, the bright-field imaging, the quantitative phase imaging as well as cell counting, profile analysis, three-dimensional (3D) imaging and differential interference contrast (DIC) imaging can be realized. With its high-resolution based computational microscopy interface, this system can be also adaptively used for telemedicine applications and point-of-care testing (POCT) in resource-limited environments.

The wide-swath greenhouse gas monitor is a hyperspectral resolution push-broom imaging spectrometer, which is used to detect global distribution of concentration of CO2 and CH4 in atmospheric greenhouse gases . Instrument Line Shape (ILS) is one of key parameters. Both radiometric calibration and spectral calibration need to convolve the reference spectrum with ILS. Therefore, the calibration accuracy of ILS directly affects radiometric accuracy and spectral accuracy. In order to accurately calibrate ILS, a test device with a tunable laser and integrating sphere was built; nine areas of the focal plane array detector were selected for the ILS calibration test, and the test results were Gaussian fitted. The results show that: taking the 1.61μm channel as an example, the accuracy of ILS is better than 1%; the spectral resolution, which is the full width at half maxima of ILS, is 0.07nm, which meets the need of retrieve applications.

A method aiming at depth of field of light field based Fresnel hologram is proposed, which is generated by using the weighted type smart pixel mapping. The conventional integral holograms are generally generated through Fourier transforming the elemental images (EI) of II into hogels. However, the non-uniform compression along optical axial is inevitable within the generation of integral hologram. The weighted type smart pixel mapping is introduced to enhance the depth of field of light field-based Fresnel hologram while keeping a compact image size and being free from moving the lenslet array. The processed hologram can reconstruct the 3D image with the combined light fields of all the integral hologram sequences. Finally, the experiment of real 3D object is numerically and optically conducted. The high matching results among them confirm this work a better performance over the conventional methods.

We present a wavelength-scanning-based lensfree microscopy that generates high-resolution reconstructions from undersampled raw measurements captured at multiple wavelengths.The reconstruction result of the standard 1951 USAF achieves a half-pitch lateral resolution of 775 nm, corresponding to a numerical aperture of ∼ 1.0, across a large field of view (∼ 29.85 mm²). Compared with other super-resolution methods such as lateral or axial shift-based device and illumination source rotation design, wavelength scanning avoids the need for shifting/rotating mechanical components. This wavelength-scanning super-resolution method would benefit the research and development of more stable lensfree microscopy system.

Fringe Projection Profiler (FPP) is one of the most popular optical non-contact three-dimensional (3D) shape measurement technologies and a research hotspot in 3D face recognition. However, due to the use of a digital light projector (DLP), high-speed, low-cost, and small-size face measurement is still a big challenge. In this article, we developed a high-speed 3D face measurement system based on near-infrared, and proposed using a projection module to replace DLP for projection. Through the accuracy experiment, the measurement accuracy of this system is better than 0.15mm. Experiments on real faces show that this method can collect 3D face data at a lower cost and at a higher speed, and meet the requirements of the field of 3D face recognition.

Conventional spectral imaging contains two-dimensional and spectral-dimensional information, but lacks spatial depthdimensional information. The light field imaging technique can acquire both intensity and direction information of the light simultaneously, so that the depth information of the target can be extracted and processed. This paper applies light field imaging technology to spectral imaging, builds a spectral light field imaging system with spectral range in the visible band and resolution of 10 nm, and proposes the theory of recovering true color images from spectral dimensional images of the same viewpoint and then performing depth estimation, and realizes an adaptive depth estimation algorithm for multiinformation fusion. The algorithm uses the correspondences information of the angular entropy representation, the scattered focus information of the color similarity constraint to obtain the initial depth and confidence level, and then fuses with the edge information to achieve the depth estimation of the scene. The experimental results show that the algorithm can solve the occlusion problem well and the edge information of the depth map is preserved intact. The theory and algorithm provide new ideas for the three-dimensional modeling of spectral images and lay an experimental foundation for the extension of spectral imaging technology to three-dimensional space and its wide application in object recognition and classification, three-dimensional display, biomedicine and other fields.

In fringe projection profilometry (FPP), the phase-shifting profilometry (PSP) is widely used in various fields such as industrial quality check and biomechanics. Among the methods of PSP for absolute phase retrieval, stereo phase unwrapping method, as a new-fashioned method, can eliminate the 2π phase ambiguities and obtain a continuous phase map with a stereo vision system, retrieving the correct fringe period with geometry constraints and depth constraints and not needing to project any additional patterns. However, ensuring the stability of stereo phase unwrapping, the frequency of fringe patterns should be limited at about 20, which affects the accuracy of the 3D shape measurement. Furthermore, in order to enhance the performance of stereo phase unwrapping, stereo block matching based on the speckles as additional auxiliary information is used, which can distinguish several periodic candidate points but still not avoid mismatches, especially for the areas of the dark regions and the object edges where the fringe quality is low. To solve these problems, we proposed a robust absolute 3D measurement using stereo cost-volume filtering for fringe orders in this paper. In the proposed method, besides the stereo block matching, a matching cost filter is designed to aggregate the preliminary matching cost for the final cost which is used to obtain the fringe order map by WTA and then the absolute phase map. Then the more robust and more accurate 3D measurement is realized compared with traditional methods. The experimental results demonstrate that the proposed method can achieve robust absolute 3D measurement using stereo phase unwrapping.

Fringe projection profilometry (FPP) has been widely applied in three-dimensional (3D) measurement owing to its high measurement accuracy and simple structure. In FPP, how to effectively recover the absolute phase, especially through a single image, has always been a huge challenge and eternal pursuit. The frequency-multiplex methods can maximize the efficiency of phase unwrapping by mixing the multi-frequency information used to eliminate phase ambiguity in the spectrum. However, spectrum aliasing and the resulting phase unwrapping errors are still pressing difficulties. Inspired by the successful application of deep learning in FPP, we propose a single-shot frequency multiplex fringe pattern for phase unwrapping approach using deep learning. Through extensive data learning, the properly trained neural networks can directly learn to obtain spectrum-aliasingfree phase information and robust phase unwrapping from single-frame compound input. Experimental results demonstrate that compared with convenient frequency-multiplex methods, our deep-learning-based approach can achieve more accurate and stable absolute phase retrieval.

We present a miniaturized microscopic imaging system to achieve multi-contrast label-free imaging. In our imaging system, a highly integrated optical system using a miniaturization lens with fixed focal length replaces the complex optical path of traditional microscope systems, significantly reducing the size of the microscope to 14*16.5*20 cm³ . A programmable full-color light emitting diode (LED), which is controlled by independently designed operating software, is used to illuminate the sample in different illumination patterns. In addition, we developed a QT-based operating software to implement the synchronous control of the hardware system to achieve various label-free imaging approaches, including bright field, dark field, rainbow dark field, Rheinberg, differential phase contrast, and quantitative phase imaging. All the microscopic imaging approaches and system parameters can be controlled by the software without any hardware modification. In addition, some functions of the cell analysis are also added to our system, which can realize cell counter, 3D information measurement. We demonstrated our miniaturized system and its imaging results through experiments, which showed that our system is simple and fast to operate, providing more visual imaging results and high-resolution 3D information for samples. The experiment on living cell demonstrated that our miniaturized system can be built into an incubator for imaging and analysis of living cells in an environment suitable for cell growth.

Lensless imaging technique, as a newly developed microscopic imaging method, combined with its corresponding image restoration algorithm, and it can obtain large-field, high-resolution three-dimensional images without labeling. Therefore, the lensless imaging system has the advantages of low cost, good portability, large field of view and high resolution. However, due to the limitation of pixel size, hardware implementation and post data processing, the imaging resolution of lensless microscopy is far away from theoretical performance. In order to solve above problems, this paper proposes a method to improve the capability registration and information coupling based on multi-wavelength illumination. Moreover, this work combines a stack of captured low-resolution images into a high-resolution result image, and the pixel super-resolution can be realized throught the reconstruction algorithm based on the exist illumination light source, Finally, experimental results utilizing USAF target demonstrate the success of proposed lensless imaging method.

Convolution neural network has been successfully applied to the super-resolution method of the visible image. In this paper, we propose an infrared image super-resolution imaging algorithm based on auxiliary convolution neural network, which uses the detail information provided by the visible image under low-light conditions for super-resolution imaging of infrared image. In this algorithm, infrared image and visible image are input into the convolution neural network at the same time to obtain high resolution infrared image. The results show that the super-resolution infrared image has more detailed information. Compared with other super-resolution methods, the proposed network can obtain the high super-resolution reconstruction efficiency.

A commercialized digital holographic microscope with complete software supporting is composed of software and hardware, and also has quantitative phase recovery and phase compensation algorithm support. The calculation speed can reach thirty frames per second, and the resolution can reach 0.775 μm under a 20 times 0.4 NA objective lens, which can realize subcellular three-dimensional reconstruction and thickness distribution measurement of transparent samples in real-time and high resolution. This paper describes the optical path geometry, algorithm flow and image calibration of the system, including the phase aberration compensation algorithm based on principal component analysis (PCA), unwrapping algorithm based on reliability analysis and subpixel displacement technology. These algorithms are processed by software in real time and can be adjusted flexibly. In addition, the whole system is tested by using phase resolution board. Quantitative phase recovery, three-dimensional structure display and differential interference contrast (DIC) display were performed for human cervical cancer cells and human oral epithelial cells, and quantitative profile measurement was conducted for pollen.