The 241-element deformable mirror model is built using COMSOL Multiphysics, and the influence function data of the driver of the deformable mirror are acquired by calculation. Based on the sub-aperture slope method, a procedure is compiled, by which the optical aberration of the 241-element adaptive optics system is corrected. By taking the root mean square value of the residual fitting error of a 3-60 order Zernike polynomial, the optimal parameter is determined by analyzing the influence of the structural parameters of the deformable mirror on the correction capability. Under this optimal parameter, the coupling factor of the deformable mirror is calculated as 11%. The influence of the system on the fitting ability of the Zernike polynomial is analyzed considering that there exist different translation and rotation errors between the Hartmann sensor and the deformable mirror in the 241-element adaptive optics system. The results show that the translation and rotation errors are not greater than 3 mm and 6°, respectively.

We proposed a fingerprint matching and localization algorithm based on orthogonal frequency division multiplexing (OFDM) modulation for indoor visible light communication known as FMLA-OFDM. The experimental results show that compared with the traditional centroid localization algorithm, the average positioning accuracy of the proposed algorithm and the positioning stability are improved by 56% and 55%, respectively. Further, with the increase of the horizontal distance between LEDs, the positioning accuracy of the proposed algorithm is obviously superior to that of the positioning model based on the least square method. Finally, the average positioning accuracy and the positioning stability of the proposed algorithm are 17% and 21% higher than those of the positioning model based on the binary on-off keying (OOK) modulation mode, respectively.

The combined effect of atmospheric attenuation, turbulence, pointing errors, and spatial correlation on the bit error rate (BER) of an atmospheric wireless optical multiple input multiple output (MIMO) system is investigated. The maximum likelihood detection criterion of the pulse position modulation (PPM) under the combined effect is derived via a Poisson counting model and an exponential correlation model. On this basis, the upper bound of the BER of the atmospheric optical MIMO system is derived via the Wilkinson approximation method. In addition, a simulation experiment is used to further analyze the influences of the various factors in the combined effect on the BER of the atmospheric optical MIMO system. The results show that when the atmospheric attenuation is constant, the spatial correlation has the most serious impact on the BER of the system, followed by the jitter error and the atmospheric turbulence. In contrast, the transmission distance is the most important factor influencing the BER of the system when the atmospheric attenuation is changeable. In addition, the performance improvement of the system resulting from the spatial diversity only works well when the channels are independent, and this improvement is not obvious when the channels are correlated.

An improved method for nonnegative matrix decomposition and endmember extraction is proposed based on hyperspectral data simplification. Further, the homogeneous regions of images can be identified by calculating and comparing the spectral information entropy of various regions. Only the most representative pixels in the homogeneous regions are selected for application in the subsequent nonnegative matrix decomposition algorithm, which considerably reduces the amount of computation required in the endmember extraction algorithm. The experimental results show that although the mean values of the spectral angles of several kinds of minerals extracted using the nonnegative matrix factorization algorithm before and after data simplification are equal, the operation time of endmember extraction after data simplification is reduced by approximately 4/5, and the operating efficiency of the algorithm is improved.

A point cloud simplification method is proposed based on k-means clustering. Compared with the bounding box method with a similar compression rate, the k-means clustering method can preserve the details better, and the result is more consistent with the dense and sparse distribution of the original data. Moreover, the surface of the constructed model is smoother.

We propose a real-time image semantic segmentation network model, which is named as Atrous-squeezeseg. Under the condition that the minimum parameter of the model is 2.1×107, the operation frame rate is 45.3 frame/s, and the pixel point accuracy and mean intersection over union can reach 59.5% and 62.9%, respectively. At the same time, in the embedded device NVIDIA TX2, the operate frame rate is up to 8.3 frame/s. The experimental results show that, compared with other segmentation algorithms, the speed and parameter quantity of the proposed model are increased.

This study proposed an image denoising algorithm based on deep learning. The scale-perception edge-protection filter was used to decompose the noise image in multiple scales. Small features, such as the image noise, were removed via scale sensing and edge preserving, and the edge details were kept unchanged. A trained convolutional neural network model was used to gather detailed information about the image, and the image was then processed using the scale-perception edge-protection filter for detail recovery. The results show that the proposed denoising algorithm can effectively reduce noises and well retain high-frequency information. Moreover, the fusion results correlate well with human visual observations.

The characteristics of fractional differential and its mask operator are investigated, and a new enhancement algorithm of fractional differential images is proposed based on local binary pattern variance (LBPV). The LBPV theory is used for the feature extraction of images. A more effective fractional mask template is constructed. The experimental results show that compared with the existing enhancement algorithms of fractional differential images, the proposed algorithm performs better in the textures and details of enhanced images.

The prominent features of the salient region are determined by focusing on the regional boundary and the object's edge pixels. Further, the uniqueness of the salient global color is used to calculate global features. Finally, the salient region is obtained using the convolutional neural network (CNN) model based on the regional and global salient features. Adopting a circular structure network is critical to eliminate the noise characteristics by referring to the surrounding environment information for multiple times. The proposed algorithm is tested using the image libraries of MSRA and ECSSD and it is found that its harmonic mean and average error associated with the average precision and recall are better than those of the current popular algorithms.

A robust affine iterative closest point algorithm is proposed based on a control point consistency constraint. The proposed algorithm establishes an objective function that is constrained by the control points, and the data point set can be affinely registered to a target point set. Furthermore, the nearest point correspondence is established between the data point and target point sets; subsequently, a random sample consensus method is used to select the high-precision shape control points, and the new affine transformation is obtained using an objective function under a control point constraint. The experimental results demonstrate that the accuracy and the robustness of the proposed algorithm are improved significantly when compared with those exhibited by the conventional image point set affine registration algorithms.

A new concept of point-cloud feature detection operator called linear intercept ratio feature detection operator was introduced in this paper. Herein, the linear intercept ratio was defined based on the geometric relation between the adjacent points to construct the function of feature point extraction modified by the Gaussian function of the distance between the points. The experimental results denote that the proposed method exhibits a decreased error recognition rate with the increase of the noise intensity in the model. Further, the proposed method can rapidly and accurately screen out the feature points, and exhibits a good anti-noise capability and an enhanced feature recognition ability.

The edge-based finite element method is introduced to systematically investigate the transmission characteristics in an antipodal overlapping finline loaded with right- and left-handed materials. Moreover, the influence of the thickness of metallic fins on the characteristics is also discussed. In addition, the paper discusses the influences of materials on the cutoff wavelength of the dominant mode, the single-mode bandwidth, the dispersion characteristics and field patterns of the dominant mode and the first higher-order mode. The results show that the cutoff wavelength of the dominant mode when the left-handed material is loaded is approximately less than that when the right-handed material is loaded. Moreover, the cutoff wavelength of the dominant mode and the single-mode bandwidth vary inversely with the expansion of the loading area, and the field line distribution of the dominant mode in the loading area is entirely different from that of the first high-order mode. However, the variation in the dispersion characteristics is similar. The numerical results provided herein are considered usefully in the design of new microwave and millimeter-wave devices.

Herein, an eight-wavelength photonic-crystal wavelength demultiplexer was designed based on the superior frequency-selection characteristics of a photonic crystal resonator. The energy band structure of the photonic-crystal with a specific lattice constant and radius was obtained using the plane wave expansion method. The law of variation in the coupling frequency of a micro-cavity was analyzed via the finite difference time domain method. The resonant coupling between the ring resonator and the micro-cavity was demonstrated. The device primarily comprised four photonic-crystal ring resonators and eight micro-cavities with different sizes. It achieved wavelength-division multiplexing at eight wavelengths of 1.37, 1.39, 1.42, 1.44, 1.50, 1.51, 1.53, and 1.55 μm. The results show that the aforementioned eight wavelengths can be output from a specific port, and an output efficiency of more than 97% can be achieved by adjusting the radii of the central rods of the micro-cavity and the peripheral dielectric rods. The designed device can be applied in integrated optics owing to its small physical size of 23 μm×18 μm and high coupling efficiency.

The ANSYS software was used to perform the numerical simulation of laser cladding. The temperature field was determined under different technological parameters. An experiment involving laser cladding of 304 stainless steel on 27SiMn steel surface was conducted under the same conditions as those in the numerical simulation. The microstructure of the cladding layer was observed using an optical microscope. The results show that the peak temperature in the molten pool increases as the laser power increases, and decreases as the scanning speed increases. The geometrical dimension of the molten pool obtained via numerical simulation is consistent with that obtained in the experiment under the same conditions. Under the laser power of 2500 W and scanning speed of 13 mm/s, the cladding layer has relatively small grains and compact structure, and shows optimal formability.

The 808 nm high-power semiconductor laser bars are used for packaging to optimize the two important factors that affect the package quality: the furnace tube temperature and the soldering time. The results show that the solder layer has the least voids when the soldering time is 100 s at a tube furnace temperature of 650 ℃. Moreover, the obtained semiconductor laser bar has the lowest smile effect value, the smallest threshold current, a more stable wavelength, and the best soldering quality.

A K-means clustering method is proposed based on Gaussian map. First, we search the neighboring points of the target points using a k-nearest neighbor search. Gaussian map is then performed on the normal vectors of the set of triangles consisting of the target point and its neighbors. The silhouette coefficient is selected as the cluster validity index to determine the optimal cluster number. According to the clustering distribution of different surfaces, the feature lines of the three-dimensional laser point cloud model are obtained. The experimental results show that the proposed evaluation index is easy to use and has less noise than other indexes. It can extract the feature lines of regular and irregular point clouds completely and efficiently.

By the gelcasting process, the porous graphite preform prepared by selective laser sintering is mixed with mullite ceramic slurry, and after freeze drying and sintering, the graphite/mullite composite is obtained. The performance of each sample is tested and analyzed, and the suitable sintering process is determined. The results show that the open porosity, density, thermal conductivity, and bending strength of the porous graphite preform are 60%, 0.518 g/cm3, 1.01 W·m-1·K-1 and 1.7 MPa, respectively. For the sintered mullite ceramics, the relative density is 80.4%, the compressive strength is 28.3 MPa, the bending strength is 27.5 MPa, and the thermal conductivity is 2.35 W·m-1·K-1. The post-processing of porous graphite preforms can avoid the occurrence of cracks in the combination zone and the poor combination of two phases, and the integrity and interfacial bonding of composites are greatly enhanced.

A machine-vision based defect detection method for packaging bags is proposed. Considering the defect detection of ice-lolly bags as an example, five kinds of eigenvalues of length, width, area, filling degree, and location relationship between the monitoring frame and the internal target region are extracted. After defect detection and classification, the following four types of defects are outputted: continuous bags, dimension errors, foreign matters on packages, and motion of packaging layout. The experimental results demonstrate that the recognition success rate of the proposed algorithm can reach 98.75%, which meets the requirements of high speed, high precision, and real time in the production process. The algorithm has been applied to an actual production line and has achieved good results.

A hue contrast enhancement algorithm based on gene expression programming optimization is proposed. A number of low illumination images are selected as the reference images and the results are compared with those of the adaptive histogram equalization, homomorphic filtering, multiscale Retinex enhancement, and color-restored multiscale Retinex enhancement. The average values of the peak signal-to-noise ratio, structural similarity and quality index based on local variance of the proposed algorithm are 25.93, 0.75, and 0.87, respectively, which are better than the other algorithms. Subjectively, the brightness and contrast of images processed by the propose method are more natural and more in line with human visual characteristics. The proposed algorithm can be widely applied to the field of machine vision in low illumination environments.

A design method for an interferometric phase imaging system integrated with optical scattering information is proposed. Three types of cell models, i.e., homogeneous cells, monocytes, and binuclear cell models, are established. The morphological characteristics of a sample can be extracted quickly and quantitatively by analyzing the phase distribution information of the sample. Thus, the substructure characteristics of the sample can be determined accurately. The necessity and advantages of combining scattering information are demonstrated through experiments on the red blood cells and neutrophils. The research results establish the feasibility of the proposed system. Additionally, the direction for improvement is proposed. The system has a simple structure and is easy to operate, and it can provide a reference for the rapid, non-destructive detection of morphological and structural characteristics of cells.

Three optimization algorithms of generalized Lagrangian multiplier (GLM), interior point and modified sequence quadratic programming are proposed. By adopting a positive-negative-positive-negative-positive lens structure with a suitable aberration distribution and setting the reasonable and generally applicable parameters, the Gaussian structure of a full frame zoom lens with zoom range of 30-100 mm, constant F-number of 2.8, field-of-view angle of 24.421°-71.608° is obtained by the proposed three methods. These three optimization methods are analyzed and compared. The actual lens group is used to replace the Gaussian structure obtained by the GLM algorithm. The full spherical design is adopted and the obtained optimal design of the full-frame zoom lens meets the imaging requirements. Three zoom and compensation curves have good linearity, and the average value of the modulation transfer function of system is greater than 0.1 at 50 lp/mm. This full spherical design provides a good initial structure for the later aspheric optimal design.

In this study, we propose an improved parallel microring antenna array with increasingly wide rings, and we also study the switching and routing methods for the filter channels. Herein, we achieve flexible switching and routing of the filter channels of the parallel microrings. Further, the transmission characteristics of the parallel microring arrays with and without the proposed improvement are analyzed using the transfer matrix method. The routing method of the filter channel of the improved three-microring array is then numerically analyzed using MATLAB. The results denote that the filter channel is closed and the light with the corresponding wavelength cannot pass when two adjacent microrings satisfy the condition of destructive interference; however, when two adjacent microrings satisfy the condition of constructive interference, the filter channel is opened, and the light with the corresponding wavelength can pass. Finally, a 2×2 router is designed using four improved microring arrays.

A double-frame digital-analog-hybrid scanning strategy is proposed, and the part of digital pulse width modulation is optimized. The driving circuit of an organic-light-emitting-diode (OLED) -on-silicon microdisplay is designed, which includes the pixel driving circuit. The scanning efficiency can reach 99.22% and the data transmission frequency can be reduced to 23.328 MHz under the conditions of frame rate of 60 Hz, gray level of 256, and resolution of 1920 pixel×3 pixel×1080 pixel. The results show that this pixel driving circuit can effectively reduce the leakage current, guarantee the gray precision, reduce the minimum current, and improve the contrast of the OLED-on-silicon microdisplay.

A low profile linear reflective polarization conversion surface in the Ku band (12-18 GHz) is proposed using a metasurface, which can convert a linearly polarized incident wave into its orthogonal polarization states in the Ku band. The principle of designing such a polarization conversion surface is given. The physical working mechanism of a polarization conversion surface is clarified. The simulation is used to determine the positions of resonance points of the polarization conversion surface. The test results show that the proposed polarization conversion surface has a conversion efficiency of more than 80% within the working band, and has the advantages of high conversion efficiency, small size, small thickness, and simple structure.

The interactions between galectin-3 and two types of peach-gum polysaccharides with different molecular weights (PGP-1 and PGP-2) were detected herein via self-assembly surface plasma resonance (SPR) imaging based on digital holography. Different concentrations of peach-gum polysaccharides and Galectin-3 were simultaneously detected on an SPR biochip prepared for detecting the concentrations. The standard curves were derived and the binding equilibrium constants of the reactions were calculated. The results show that the two types of peach-gum polysaccharides can directly bind to Galectin-3. The binding equilibrium constants of PGP-1 and PGP-2 are 8.36×105 and 1.24×105 M-, respectively. The binding curves conform to the law of biomolecular interaction, demonstrating the feasibility of the proposed method in high-throughput biological detection. The proposed method can be easily controlled and is simple, label-free, and inexpensive. It is potentially applicable to the high-throughput microanalysis technology.

A portable polarization-entangled quantum photon source with an optical output as well as an auxiliary electronic output in the 1550 nm telecom band is developed based on spontaneous four-wave mixing in a dispersion-shifted fiber. The polarization-entangled photon pairs of signal and idler waves with wavelengths of 1558.17 nm and 1548.51 nm are outputted from the standard single-mode fiber port, respectively. The electrical signals with tunable delay and frequency division are outputted from the coaxial cable port, which are used for the single photon detectors and the data acquisition system. The results show that when the photon generation rate in a single channel is about 0.0007 pair/pulse, the ratio between the coincidence count rate and the accidental coincidence count rate is up to 22.8, and the visibility of polarization-entangled interference fringes is 87%.

The formation mechanism of a femtosecond-laser plasma channel is mainly discussed. The research progress on electromagnetic wave transmission in a femtosecond-laser plasma channel is reviewed. The plasma channels are classified into three types: single-channel transmission line, double-channel transmission line, and cylindrical hollow waveguide. Finally, the development trend of electromagnetic energy transmission via femtosecond laser plasma channels is prospected.

Biospeckle is an optical non-destructive testing technology that uses laser refraction and reflection inside the object to reflect its internal information. Biospeckle imaging equipment and image processing algorithms are constantly improved, and the application fields are gradually expanding. Due to the existence of interference factors, the precision of modelling is still the focus of researchers at present. The speckle image processing algorithm is reviewed in detail, and the application of biospeckle technology in fruit quality detection is investigated. The main imaging equipments are summarized and suggestions for improvement are proposed to provide inspiration for future research.

In this paper, we discuss the optical properties of all-dielectric nanoparticles in detail, introduce the preparation methods of all-dielectric nanoparticles, and analyze the advantages and disadvantages of various preparation methods. Further, we review the applications of all-dielectric nanoparticles in the fields such as high-index nanometer resonators, nano-antennas, metamaterials and metasurfaces, and nonlinear nanophotonics. Finally, the research focus and development direction of all-dielectric nanoparticles are proposed.

The latest research results and applications of superpixel algorithms and evaluation indexes are summarized. Many superpixel methods are compared by using the evaluation indexes such as boundary recall, under-segmentation error rate, and compactness. The corresponding advantages and limitations are also analyzed. The experimental results show that the current superpixel methods are greatly superior to the previous methods in terms of accuracy and efficiency, and the applications of superpixel algorithms are growing constantly. However, it remains difficult to satisfy the requirements of the superpixel performances in some special applications. Therefore, it is necessary to develop the new methods that are more robust and have better adaptability.