To tackle the influence of a light source in the field of view on imaging, this paper presents a new underwater image restoration method suppressing its interference. The underwater imaging model considering the influence of a light source is first established, and then the light source radiation is separated depending on the different gradient distribution between the image interfered by the light source radiation and the image without the influence of light source. With regard to the acquired image after the interference of light source is suppressed, its global background light and the underwater transmittance are estimated by the maximum filter and the median filter, respectively, which effectively reduces the influence of the light source in the field of view on parameter estimation. The experimental results verify the effectiveness of the underwater image restoration method proposed in this paper, which can significantly improve the visibility of underwater scenes. In addition, this method can also achieve a good effect of underwater image restoration without the influence of light source.

Aiming at the problem of ground-space slope reverse modulation wireless optical (FSO) communication under the influence of atmospheric turbulence, the optical intensity scintillation under the reverse modulation slant path link is studied based on the three-layer height spectrum model, and the probability density function expression of the atmospheric turbulence fading coefficient under weak turbulence is derived based on this formula, the average outage probability, average bit error rate and average channel capacity of the system are deeply studied, and the influence of factors such as zenith angle, turbulence intensity and receiving aperture size on the communication performance of the system are analyzed. The research results show that in the case of low signal-to-noise ratio and long-distance communication, the receiver aperture is appropriately increased, and the ground-space reverse modulation slant path wireless optical communication system can still maintain good communication performance. With the decrease of the zenith angle and the refractive index structure constant of the near-earth atmosphere, the influence of turbulence on the slant path communication link is obviously weakened.

To make clear the generation mechanism of microbending loss in optical fibers and effectively estimate and test microbending loss, this paper analyzed two mechanisms, i.e., microbending caused by pure bending and that by coupling, briefly described the most relevant and effective theoretical formulas, and pointed out the most important factors affecting the microbending loss. Two methods to test microbending loss were proposed for the first time, respectively using an optical fiber reel covered by metal mesh and a double-slot plate. In light of the transverse strain and pure bending theory, two corresponding simulation algorithms were established respectively. Depending on the decomposition and simulation of microdistortion and microbending of the tested optical fibers in test platforms, the microbending loss was estimated. The established platforms were then used for testing. The verification results show that the calculated values of the proposed simulation algorithms are in good agreement with the measured results, and the proposed test methods can be employed for standard formulation and extended.

This paper proposes a high sensitive fiber optic microphone based on a micro-electro-mechanical system (MEMS) corrugated diaphragm. The Fabry-Perot interference cavity is obtained with the corrugated diaphragm, which can further improve the sensitivity of the fiber optic microphone. In the theoretical analysis, the finite element analysis software COMSOL is used to model the diaphragm, and the effects of the corrugation width, corrugation depth, and the separation between two corrugations on the sensitivity and working frequencies of the diaphragm are analyzed by simulation. The experimental results show that the sensitivity of the fiber optic microphone based on the MEMS corrugated diaphragm is 630 mV/Pa, and the frequency response range is 50 Hz--8 kHz; and the structure of the fiber optic microphone is simple, which can be well used in the fields of far-field speech recognition and so on.

Stimulated Raman Scattering (SRS) is the primary restrictive factor in the improvement of output power and stability of high-power fiber laser. Effective filtering and suppressing of SRS are a key dilemma that need resolution. To effectively filter the Stokes light caused by SRS, long period fiber grating (LPFG) can couple core mode with special wavelength to cladding modes to realize bandwidth filtering. Hence, the CO2 laser beam combined with scanning galvanometer is used to fabricate LPFG on double cladding fiber to filter and suppress Raman signal in high-power optical fiber oscillator and an evident inhibition is achieved. Results show that the filtering ratio of Stokes light beyond 15 dB, the purity of signal laser and stability of the system have improved considerably. Because there is no effect on fiber core, theoretically, the proposed method has a very low signal optical transmission loss, which is of great value in high-power fiber laser.

Loss and dispersion characteristics of dielectric-coated metallic hollow waveguides are simulated for frequencies ranging from 140 GHz to 220 GHz. Results show that the dielectric layer coating cannot reduce the transmission loss in this frequency band when the operation wavelength is close to the inner diameter of the waveguide. However, the electric field profile of the model is distorted because the dielectric layer and dispersion of the waveguide switches from positive to negative. The dispersion variation and zero-dispersion position in dielectric-coated metallic hollow waveguides with different dielectric layer thicknesses are analyzed and summarized. Dispersion compensation methods and schemes for metal waveguides are proposed using waveguides with different dielectric layer thicknesses. The result shows that zero-dispersion can be achieved at different frequencies and flat and low dispersion over specific frequency bands can be realized. It guides structural design and fabrication parameter control for the waveguides used in this frequency band.

When analyzing the influence of proportional gain coefficient of the loop filter in the clock recovery algorithm on system bit error rate, we find that the bit error rate of the signal first decreases and then increases with the increase in proportional gain coefficient, and the optimal proportional gain coefficient varies with received power. In response to this problem, we propose a full-digital clock recovery algorithm having a full real-time dynamic proportional gain coefficient based on Gardner algorithm to deal with power jitter channels and improve the sensitivity of the algorithm. The algorithm is then verified by online experiments in a 2.5 GBaud coherent communication system built upon the Altera Stratix-V 5SGXMA7K2F40C3 FPGA platform and quadrature phase-shift keying modulation. With the forward error correction threshold of 2×10 -4 in KP4 as upper limit, the dynamic parameter algorithm can reduce the minimum received power for the system to reach this threshold from -47 dBm of the fixed parameter algorithm to -49.5 dBm when the power jitter frequency is low. When the frequency is high, the lowest power and frequency of the jitter that the fixed parameter algorithm can withstand are -46.5 dBm and 100 Hz respectively, whereas those of the dynamic parameter algorithm are -50 dBm and 1 kHz respectively.

In this paper, a beak-shaped optical tweezer with annular core fibers is designed and studied by theoretical simulation. The finite element method is used to simulate the optical field intensity distribution of the optical tweezers, and the optical field images of the optical tweezers under different bending conditions are compared. The results show that bending deformation increases the light intensity of the side evanescent field. The Maxwell stress tensor method is used to analyze and calculate the trapping force of the optical tweezers for particles. The new tweezers are compared with the conical optical fibers to analyze the relationship between the bending radius and the trapping force, as well as between the particle radius and the trapping force. The results show that the designed optical tweezers can not only capture particles at the tip but also capture and transport particles at the side. The new optical tweezers can be applied to auxiliary research on cell biology.

To solve the problems that the color of colonic polyps is similar to the background color, and the different sizes of polyps affect the segmentation effect, this paper proposes an improved image segmentation network for colonic polyps with a double U-shaped structure. The proposed model is based on the DoubleU-Net architecture. Firstly, the spatial attention block is integrated into the skip connection of the U-shaped structure to extract the correlation information of spatial features. Secondly, the channel attention block is incorporated into the skip connection to clearly express the dependency of useful channels and suppress the features unrelated to the polyp segmentation task. Finally, at the input of the encoder with the second U-shaped structure, the selective kernel block is introduced to adaptively select different receptive fields and improve the target segmentation accuracy. The experimental results show that the proposed method is better than the existing methods in terms of objective indices and visual effects. The research results can provide new reference for the early detection and surgical planning of colonic polyps.

In order to overcome the loss of detail information caused by down sampling of traditional liver tumor segmentation networks and extract rich multi-scale information at the same time, this paper proposes an algorithm of liver tumor segmentation based on dilated convolution of stacked tree aggregation structure. First, a residual dense module is proposed in the encoder network. Then, a dilated convolution module of stacked tree aggregation structure is added to the encoder-decoder network, which can effectively eliminate the checkerboard artifacts caused by ordinary dilated convolution and improve the segmentation accuracy. Finally, a weighted loss function is used to solve the problem of the imbalance between the foreground and the background in the image. The experimental results show that the Dice similarity coefficient, pixel accuracy rate and intersection ratio of the algorithm on the computer tomography image data set are 0.8026, 0.7974 and 0.7317, respectively.

Combining multiple pushbroom cameras along the sampling direction can overcome the limitation of the detector scale, and effectively expands its field of view and improves the imaging efficiency. Existing stitching methods are based on strict geometric calibration of each sub-field, and strongly dependent on expensive calibration fields. This paper presents a relative orientation method of sub-field of view based on sparse ground control points. This method establishes a relative orientation model with tie points and digital elevation model. To ensure the stability of the solution, it is supplemented by the absolute constraints of the sparse control points. Relationship between adjacent cameras can be achieved by this method, and simulation data was used to verify the method. Results show that the relative orientation parameters can support seamless stitching of sub images. The accuracy of image stitching is better than 2 pixel, and the plane error of geometric positioning accuracy is better than 3 pixel.

Aiming at the shortcomings of traditional laser resonator composed of flat or spherical mirrors that require high precision adjustment and allow a small rotation misalignment angle, the Luneburg lens and spherical total reflection surface are combined as beam retroreflective devices. We performed numerical simulations on the reflection characteristics of the retroreflector under different incident angles of Gaussian beams and different distances between the central axis and the optical axis of the Luneburg lens. The results demonstrate that even when the linearity is as small as one-tenth of the wavelength of the light wave, the reflectors can still achieve good directivity and retroreflection with stable reflected amplitude for incident Gaussian beam distributed in a large range in the angular and lateral direction. Therefore, it is theoretically feasible to use it as a laser optical resonator mirror, and it has the advantages of low adjustment requirements, large cavity length lateral, and angular movement range. And the size of the retroreflector is small, which is easy to realize the miniaturization of the resonator mirror.

A scanless and lensless 3D endoscopic imaging method based on a coherent fiber bundle (CFB) is proposed. Compared with conventional endoscopic imaging method, the proposed method uses the end face of a bare fiber bundle as a probe and does not require integrating any low-light-level mechanical components. Therefore, the proposed method provides a simple structure, small volume, and easy miniaturization. Moreover, it combines synthetic aperture lidar (SAL), broadband linear frequency-modulated lidar, and digital holographic lidar. A broadband linear frequency-modulated laser generated an active illumination signal, and the CFB collected and transmitted the scattering echo of a target. Simultaneously, based on off-axis digital holography, a high-speed CCD array camera recorded the echo signal. Finally, SAL data processing method is used to reconstructed a 3D image of the target. The mathematical model of the 3D endoscopic imaging system is established, and the corresponding data processing method is detailed. Theoretical and simulation results show that the proposed method overcomes image pixilation caused by the inherent structure of the CFB and has the ability of precise transversal and longitudinal 3D imaging perceptions.

To tackle the problems in traditional high-precision angular displacement measurement, such as complicated manufacturing process, complex multi-probe structure and difficult installation, this paper proposes a new relevant full-circle method based on the construction of a rotating optical field, inspired by the generation principle of a rotating magnetic field. A sinusoidal uniform grating surface on the full circle is divided by an equal interval into four groups of 0°, 90°, 180°, and 270°. Alternating light intensity signals are provided for the full-circle grating surface by a one-way alternating optical field, which are transformed into electric traveling wave signals via photoelectric conversion and micro-controlled phase shift. The angular displacement measurement is realized by the interpolation of the phase difference between reference signals and electric traveling wave signals with a high-frequency clock pulse. An experimental platform is established for the comparative experiment on different principle prototypes. The influence of manufacturing precision and number of pole pairs on the measurement error is verified by experiments. The measurement error can reach ±5.0″ when the grating surface with a central angle of 2° is used in the full-circle measurement range. Since the measurement error is mainly composed of periodic error components, the measurement precision will be greatly improved after the correction of harmonic error and the quality improvement of optical field distribution. This paper verifies the suppressing effect of the proposed method on manufacturing and installation errors and provides an effective solution to the high-precision measurement for low-precision manufacturing processes.

Transparent objects such as glass and lenses are commonly used in optical systems, and their surface quality greatly affects the performance of the host systems. Considering polarization characteristics of light, surface defect detection method using polarized transmission structured light is proposed. The constructed transmission system generates fringe-encoded polarized structured light for projection onto the surface of the measured object. The structured light deforms after passing through the object. Then, the deformation fringes are collected, phase information is extracted, modulation is solved, and surface defect information is obtained from the measured object. Experimental results show that the proposed method can eliminate the effect of dust and improve the signal-to-noise ratio for detection. Therefore, the proposed method can be suitable for surface defect detection in transparent objects such as flat thick lenses and high-curvature optical lenses.

A new real-time three-dimensional (3D) measuring method is proposed based on single-shot four-grayscale fringe projection, in which different non-zero three-scale grays in the four gray levels are cyclically arranged in equal width to encode a frame of three-grayscale fringe pattern. When the three-grayscale fringe pattern is projected onto the measured object, the corresponding four-grayscale deformed pattern is captured by the imaging camera, and the fourth grayscale corresponds to the shadow area in the deformed pattern. By the image segmentation method, non-zero three-grayscale fringes in the four-grayscale deformed pattern can be extracted respectively. After binarization, three frames of binary deformed patterns with 1/3 duty cycle and relative displacement of 1/3 period can be demodulated. So three frames of sinusoidal fringes with a 2π/3 shifted phase can be extracted by the inverse Fourier transform after filtering the corresponding fundamental frequency of the three frames of binary deformed patterns, and the 3D shape of the object can be reconstructed by phase measuring profilometry (PMP). The experimental results confirm the feasibility of the proposed method. Because of the fast refreshing rate and the insensitivity to the gamma effect of the projector, the proposed method has a potential application prospect in real-time 3D measurement.

In order to meet the needs of accurate, fast and portable measurement of greenhouse gases such as CO2 and CH4, a multi-wavelength greenhouse gas measurement system based on tunable Fabry-Perot interferometer (FPI) sensor is introduced in this paper. Using FPI as wavelength selective element, the continuous measurement in band of 3100--4400 nm (wavenumber range is 3226--2273 cm -1) is realized through interference filter. The correlation coefficient is obtained by fitting the measured absorption spectrum with the absorption cross section of CO2, and then the gas concentration of CO2 can be calculated. To correct the non-linear absorption effect that caused by the low resolution of the instrument, the iterative algorithm is used to optimize the CO2 absorption cross section. The results show that the measurement error is reduced by 18% compared with the direct fitting calculation when concentration of CO2 is 4.08×10 -4. The interference optimization, precision, the limit of detection and other parameters of the system are verified. The experimental results show that the concentration of residual CO2 in the air chamber is less than 1×10 -6, the instrument precision is ±1.32×10 -6, and the detection limit is 1.13×10 -6 (2σ, 2 times standard deviation) when the time resolution is 10 minutes. Moreover, the system is used to carry out outdoor measurement in Hefei Science Island for one week, and the daily variation results and diurnal characteristics of concentrations of CO2 are obtained, which verified the stability and reliability of the system.

A grating-lobe-suppressed aperiodic optical phased array optimized by the hybrid bat algorithm is proposed to achieve a large beam scanning range. After the operators of the genetic algorithm are combined with the traditional bat algorithm, the silicon-based 64-unit one-dimensional aperiodic optical phased array is optimized at a wavelength of 1550 nm and a minimum pitch of 2 μm. The side mode suppression ratio is better than -8 dB in the 80° scanning range, and the maximum far-field beam divergence angle is 0.3°. The insensitivity of the optimized results to wavelength change is also discussed. With an 8×8 two-dimensional rectangular planar array as an example, the optimization method of the two-dimensional aperiodic optical phased array is demonstrated.

To address the problem of target loss facing existing video tracking algorithms due to high mobility of targets or rapid deformation of asymmetric rigid targets, this paper proposes a video tracking algorithm based on the correlation filtering adaptive model and the redetection mechanism for average peak-to-correlation energy (APCE). The adaptive model tracking algorithm can adjust the model in real time according to the clarity of the target area, thereby effectively ensuring the accuracy of the target tracking model. Experimental results show that integrating the adaptive model tracking algorithm into the discriminative scale space tracking (DSST) model can enhance the tracking effect of the model on highly mobile or rapidly deforming objects. While guaranteeing tracking speed, the integration also raises the average accuracy of the original DSST model by 18.3 percentage points and the success rate by 15.2 percentage points. In addition, combining the adaptive tracking algorithm with the APCE redetection mechanism can ensure the stability of the algorithm.

Controllable printing of microstructure with specific composition and structure distribution is the key technology to miniaturize and integrate nonlinear optical devices. Spatial and geometry controls of second order nonlinear optical properties at the micrometer scale have been achieved in soda lime glass (SLG) using microthermal poling with nickel mesh as anodes and SLG as cover glass. Microstructured second harmonic generation (SHG) can be correlated to the dielectric barrier discharge (DBD) of a poling sandwich among an SLG sample, a nickel mesh, and a SLG cover. Experimental results show that the DBD induces an internal electric field with a geometrical intensity distribution in the SLG sample and cover. Furthermore, energy dispersive X-ray spectroscopy and Raman spectroscopy reveal that microstructured SHG patterns are attributed to the internal electric field induced by Na + redistribution and structural rearrangement of SLG. Microthermal poling can be used to fabricate nonlinear optical components.

The focusing and leveling system is mainly realized by the measurement principle of optical triangulation. Linear charge coupled device (CCD) has the characteristics of high frame rate and high resolution, which is suitable for high-speed and high-precision triangulation measurement. A focusing and leveling project requires a linear array CCD with 2048 effective pixels and a line readout frequency of 5 kHz. For this reason, it is proposed to overclock the S11156-2048 linear array detector with a maximum rated readout frequency of 10 MHz to 12.5 MHz. The read-out drive is designed to meet the requirements of the system line read-out frequency. First, the feasibility of overclocking readout is verified theoretically. Then, the design of power supply, timing drive and analog front end is studied in detail. Finally, the photoelectric performance of the linear CCD imaging system is compared by using an integrating sphere light source at 6.0 MHz rated frequency and 12.5 MHz overfrequency. The experimental results show that the S11156-2048 detector adopts 12.5 MHz over-frequency readout design with no significant degradation in its optical and electrical performance, the signal-to-noise ratio is 54.22 dB, and the charge transfer efficiency is 0.999974, which meets the imaging performance requirements of the focusing and leveling system.

Aiming at the influence of the polarization aberration caused by the freeform surface on the imaging quality and measurement accuracy of the system, the polarization aberration analysis method of the freeform off-axis optical system with fringe Zernike polynomial is proposed based on Jones representation, and an analytical model of polarization aberration for freeform reflection optical system is constructed, and the effects of Z5 and Z6 in fringe Zernike polynomials on the polarization aberration distribution of off-axis field of view optical systems are analyzed. Through the tracing of the polarized light in the full field of view, it is verified that the freeform surface acts on the three polarization aberrations of the field of view off-axis optical system, such as phase aberration, diattenuation, and retardance. A full-field polarization aberration analysis is carried out for a large field of view off-axis three-mirror optical system with freeform surfaces. The analysis results show that the phase aberration distribution is consistent with the vector height distribution of the freeform surface, and the two-direction attenuation and phase delay are 50% of the total system.

The imaging spectrometer, which uses a multi-zone grating as the dispersing element, can realize the sharing of optical elements. It has the advantages of wide wavelength band, high spectral resolution, and compact structure. However, the stop of this system is located at the multi-zone grating and the broadband beam is incident on the whole grating area of each channel at the same time. The multi-order diffraction of the gratings with different frequencies will cause interference among different channels, and thus it is difficult to be removed by the order-sorting filter. This problem seriously influences the extraction and analysis of the effective spectral signals, and even overwhelms the effective target signals. This paper uses ray tracing to determine the source of multi-order diffraction stray light in the system, and calculates the diffraction stray light coefficient with the analysis model. According to the structure of the optical system, two methods for suppressing the multi-order diffraction stray light using the linear variable filter and the stop coating are proposed, and the bandwidth requirement of the filter and the improvement effect of diffraction stray light coefficients using the stop coating are analyzed. The analysis results show that the stop coating method can reduce the diffraction stray light coefficient to 0.57%. Compared with the linear variable filter method, the stop coating method is easier to implement, and is more preferred for the suppression of multi-order diffracted stray light in the imaging spectrometer with multi-zone grating.

A multi-layer gear-shaped metamaterial absorber with broadband and high absorption is designed based on multi-scale feature structure stacking and Fabry-Perot resonance theory. The absorber is composed of two layers of media-metal gear-shaped stacks with different sizes and the finite difference time domain method is used to analyze the absorption characteristics. Numerical results reveal that the absorptivity of this absorber in the 300-4000 nm band is above 89%, the average absorptivity can reach 94%, and the absorber shows certain polarization insensitivity. In addition, under 60° oblique incidence, it can still maintain an average absorbance of 93%. From the electromagnetic field distribution at the resonance wavelength, one can see that the broadband and high absorption characteristic of this absorber is mainly attributed to Fabry-Perot resonance, surface plasmon resonance, localized surface plasmon resonance as well as hybrid coupling among them. The proposed absorber has potential application value in photovoltaic cells, invisible devices, and other fields, due to broadband and high absorption, large-angle absorption, and polarization insensitivity.

For target tracking based on rotational double prisms, the relation between beam steering and prism rotation is nonlinear. Based on the two groups of inverse solutions of prism orientations derived using the two-step method, the directional derivative in the moving direction of the target is calculated to derive the ratio of the rotational speed of the prisms to the slewing rate of beams, which is used to discuss the requirements of tracking targets with different orientations in different moving directions for driving and controlling over prism rotation. It is found that for a given system, the requirements depend only on the deflection angle of the present target orientation and the angle between the target moving direction and the radial direction. The tangential and radial movement of the target may result in control singularity for prism rotation on the inner and outer edges of the field of regard, respectively. The requirements of target tracking for driving and controlling over the rotation of the two prisms are slightly different. There are minor differences in the analysis results based on the two groups of inverse solutions.

We experimentally investigate the gain spectra of a pair of conjugate beams by stimulated degenerate four wave mixing (DFWM) in cesium vapor. The results show that in a degenerate two-level transition system, the optical amplification effect based on stimulated DFWM can occur only when the angular momentum of ground state is greater than or equal to that of excited state. After considering all Zeeman sub-states of atoms, the qualitative theoretical analysis shows that the process is based on multiple N-type transition loops under the premise of stable ground state Zeeman coherence and large population of atoms at the initial Zeeman level. The effects of experimental parameters such as pump power, atomic number density and single photon detuning on the gain of DFWM signals are further analyzed. Moreover, it is found that the amplification efficiency of DFWM can be significantly improved by introducing an additional driving beam with a wavelength of 852 nm.

Automatic radiometric calibration is the latest development trend in the field calibration of satellites. Its core is to obtain TOA ( top of atmosphere) reflectance that can be compared with satellite observation data on a regular basis. This paper developed an on-orbit radiometric calibration method for satellites based on the TOA reflectance at the Baotou site. Firstly, the empirical model for TOA reflectance was built based on the long-time-series observation data from a high-precision reference satellite over the automatic radiometric calibration field. This model was used to correct the TOA reflectance of the automatic radiometric calibration field at other moments. Consequently, the uncertainty of TOA reflectance was reduced, and the accuracy of radiometric calibration was improved. The proposed method was validated and analyzed with Sentinel-2A/2B as the reference satellite and the TOA reflectance of sand targets at the Baotou site as the research object. The results show that the model built in this paper has high accuracy, and the average relative difference between the predicted values of the model and the observed values of Sentinel-2 is less than 2% (less than 3% in the blue band). With this model, the relative difference between TOA reflectance at the Baotou site and the actual TOA reflectance observed by Sentinel-2 can be greatly reduced and the inconsistency is decreased from 5%--6% to less than 3%. These results validate the effectiveness of the proposed method.

A novel static Fourier transform spectrometer (FTS) is proposed. The interference system of the spectrometer uses two micro mirrors with periodic structure to replace the plane mirrors in the Mach-Zehnder interference system, without any driving parts and generation of backtracking light, which realizes the static state and high optical flux of the FTS. In this paper, the working principle of static FTS is introduced, the Mach-Zehnder interference system model based on micro mirrors is established, and the interferogram and spectrogram in frequency domain of the interference system are obtained by simulation. Compared with the ideal results, the fringe contrast is reduced and the reduction in the edge area is more serious in the interferogram obtained by simulation, and the peak value of the main frequency is reduced and the baseline noise is obvious in the spectrogram in frequency domain obtained by simulation. The analysis shows that when the incident beam is inclined, the stepped periodic structure of two micro mirrors will not only cause the uneven energy distribution of the interference light field and the deviation of two coherent beams, but also cause the diffraction effect. The simulation results show that the distortion of information in interferogram can be effectively reduced by increasing the number of subreflectors of two micro mirrors and the length of subreflectors. Furthermore, the ideal spectrogram in frequency domain can be obtained by reducing the step interval between the subreflectors of the two micro mirrors under the premise of satisfying the performance of the spectrometer.

Thin film deposition is the core in the filter preparation process and its optimization is of great significance for the stability improvement of thin film filters. In this paper, a Zr thin film was deposited by pulsed direct current (DC) magnetron sputtering at different working pressures (0.05--1.0 Pa) of Ar. The surface of the thin film was measured by a Zygo interferometer and then the film stress was calculated. The microstructure changes of the thin film were characterized by an X-ray diffractometer and an atomic force microscope. The results show that the Zr thin film exhibits compressive stress when the working pressure is higher than 0.1 Pa, and the stress decreases slowly with the decrease in working pressure. Furthermore, when the pressure is 0.05 Pa, the film exhibits tensile stress. The variations in the phase structures and surface roughness are analyzed to further explain the generation mechanism for tensile stress. The results show that the main reason for the tensile stress forming is the microstructure change caused by plastic flow in the metal. Our study optimizes the deposition process in the preparation of low-stress and self-supporting Zr filters and has significant application value in extreme ultraviolet lithography, synchrotron radiation, and astronomical observation.

Under the influence of metamer mismatching, one color under a standard illuminant can become various colors under a test illuminant, which challenges color constancy phenomenon. In order to investigate the effect of metamer mismatching on color constancy in two-dimensional simulated scenes, we analyzed color constancy data obtained in asymmetric color matching performed on a CRT monitor in the context of metamer mismatching. The results showed that color constancy indices of Munsell surfaces had almost no correlations with metamer mismatching indices under each chromatic illuminant. The matched points of observers were close to the theoretical points based on traditional perfect color constancy rather than metamer mismatch centroid chromaticities. The results indicate that observers’ matching is not affected by metamer mismatching degrees of surfaces; in asymmetric color matching, observers tried to find a chromaticity point which was produced by a spectral reflectance close to that of the standard surface under a chromatic illuminant, rather than a centroid chromaticity point of metamer mismatch volume.