Low-light satellite can capture the information of town lights, fishing boat lights, and fire points to form nighttime light remote sensing images (hereafter referred to as nighttime light image, NTL image). High-resolution NTL images help to reflect human activities on the ground more clearly. It has been widely used in urban expansion, economic evaluation, carbon emission analysis, and other fields, and has become an important data source for natural and social science research. The change analysis based on multi-temporal NTL image is of great significance for monitoring human ground activities and urban development, and geometric correction of multi-temporal remote sensing image can guarantee the spatial alignment of the multi-temporal image, which is the basis for change analysis of multi-temporal NTL image. Due to the characteristics of night imaging of NTL images, there is lack of geometric reference for NTL images at home and abroad, so few scholars have carried out geometric correction research on NTL images. According to the imaging characteristics of NTL images, this paper proposes to take high-precision road network data as geometric reference, extract control points by automatic matching between NTL images and road network data, and then realize geometric correction of NTL images. Taking the Luojia 1-01 (LJ 1-01) satellite NTL image as an example, the experimental verification shows that the accuracy of geometric correction based on road network control can reach the sub-pixel level, which verifies the feasibility of the proposed method. This paper verifies the rationality of using the road network as the benchmark data for NTL images, provides a feasible idea for subsequent scholars to study the geometric processing of NTL images, and ensures the geometric quality of data for the application of multi-temporal NTL images.

Few-shot object detection is a valuable task and is yet to achieve great progress. Most research studies generate class prototypes by support samples to guide the detection. However, there still exists some challenges with this type of approach. On the one hand, the performance is highly sensitive to the quality of class prototype encoding. But limited training data impede the procession of class prototype learning. On the other hand, similar class prototypes distribute densely in the feature space, which easily leads to misjudgment. In this paper, we propose a better class feature representation (BFR) method to obtain more representative and more discriminative class prototypes with a small size of samples. First, BFR obtains the support set feature vector with richer semantic information via different feature aggregation methods. In addition, the distance metric function is used to identify the outliers and reduce the noisy interference with dynamic weights adjustment. Finally, BFR constrains the distribution of feature prototypes through distance loss, so that different prototypes are far away from each other in the feature space to reduce misjudgment. Extensive experiments on public benchmark datasets show that our method achieves superior performance with the main components bringing an 11.3% performance improvement.

Traditional bottom light emitting devices are commonly limited by the light transmittance of the substrate and electrode, while the recent emerging inorganic alternating current electroluminescence (ACEL) devices, which manifest excellent application prospects in the field of flexible display are mainly prepared by directly mixing thermosetting resin and powder materials. Here, the electrode layer, dielectric layer and light-emitting layer based on different electrode structures, were printed in different orders and structures by the inking technology. Printable inorganic ACEL devices with bottom-emission structure (BES), top-emission structure (TES) and planar electrode structure (PES) were successfully obtained. The measurements of luminescent properties of printable ACEL with different electrode structures under different voltages by spectrometer indicated that BES-ACEL had the best brightness. Besides, TES-ACEL and PES-ACEL have the outstanding advantages of no substrate limitation and interaction characteristics, respectively. Therefore, this work opens up the new path for ACEL devices towards application in the fields of printing and packaging, interactive and intelligent display.

The work aims to study the influence of the slant edge length, slant edge angle, length-width ratio and height-width ratio on the compressive strength of an octagonal box, with the length, height, width and slant edge angle set around 225 mm, 150 mm, 150 mm and 135∘ respectively. First, the single factor analysis method was used to analyze the influence of each factor on the compressive strength. Second, the parameters of each structure in the orthogonal experiment were set to conduct the proofing test. Through single factor analysis, it is found that when the variable is the slant edge angle, the compressive strength reaches the maximum when the angle is 135∘. When the variable is length-width ratio, the compressive strength reaches the maximum with the ratio of 1.3:1. When other parameters are fixed, the compressive strength increases with the increase of the slant edge length and decreases with the increase of height-width ratio. According to the orthogonal experiment, when the slant edge length, angle, length-width ratio and height-width ratio of the octagonal box are 35 mm, 150∘, 1. 2:1 and 0.6:1, respectively, the compressive strength is the maximum. Furthermore, there is a synergy between these four structural parameters; slant edge length has the greatest influence on the compressive strength of octagon box, followed by height-width ratio, slant edge angle, and length-width ratio.

CIE has recently recommended a new color appearance model CIECAM16 to replace CIECAM02. It was also intended to recommend a uniform color space (UCS) based on CIECAM16, CIECAM16-UCS, for predicting color difference. However, there was a debate as to whether CIECAM16 can be used to predict color difference, since it takes tristimulus values calculated using the CIE 2∘ observer as input, while tristimulus values for color difference are usually calculated using the CIE 10∘ observer, as in the CIEDE2000 formula. Thus, further evidence is needed before CAM16-UCS can be recommended by the CIE as uniform color space. In this paper, we investigate the likely difference in color difference prediction when using CAM16-UCS, and its potential statistical significance. Firstly, the reflectance of each pair of color samples in the visual color difference datasets: BFD-P, Leeds, RIT-DuPont and Witt was generated based on the given tristimulus values. Then, color difference ΔE_{2, F} and ΔE_{10, F}, with F being the CAM16-UCS, can be computed under any illuminant using the 2∘ and 10∘ observers respectively. Comparison results showed that the difference between ΔE_{2, F} and ΔE_{10, F} was not statistically significant. Finally, both ΔE_{2, F} and ΔE_{10, F} were used to predict the visual color difference DV and the STRESS values between ΔE_{2, F} and DV , and between ΔE_{10, F} and DV respectively. Statistical tests showed that the differences between ΔE_{2, F} and DV , and between ΔE_{10, F} and DV were not significant. Hence this study shows that CAM16-UCS can be reliably used for predicting color difference. The findings are valuable for CIE to recommend CAM16-UCS as a uniform color space. Currently CIE TC1-98 is investigating the establishment of a new colorimetric system based on cone response tristimulus values LMS. It is expected that the reflectance datasets generated from the color difference datasets will be useful for the evaluation of any new colorimetric system.

In order to realize a fault diagnosis for rolling bearings, a visualization of signals combined with proper orthogonal decomposition (POD) and continuous wavelet transform (CWT) were proposed. Thus, the problem of fault diagnosis is changed into a pattern recognition of images. With an establishment of Convolutional Neural Network (CNN) model, the fault diagnosis is completed and the main contents are as follows. Firstly, various vibration signals of faults were converted into a series of images to be distinguished with Continuous Wavelet Transform (CWT), and Proper Orthogonal Decomposition (POD) is applied to make an analysis and filter. The main principles of images were retained and reconstructed. Secondly, a CNN model is established to realize the pattern recognition of reconstructed ones. A model with two layers of convolutions and two layers of pooling was established. Then the CWT-POD-CNN model is used in two experiments of fault diagnosis successfully. Thirdly, a series of methods was used to make a comparison to prove the advantage of given work. Radial Basis Function (RBF), Back Propagation Neural Network (BPNN) and Support Vector Machines (SVM) were all combined with the same reconstructed samples. Furthermore, the CWT-CNN and CWT-POD-CNN models were also compared. On the data of rolling bearings on internet, an accuracy near to 100% was obtained in both models. As to rolling bearings in a real printing equipment, accuracy of 92.59% is obtained with constructed images and higher than 86.67% without POD. The proposed method is proven to be effective and useful in the analysis of fault signals in actual condition.

This paper describes two new interim connection spaces (ICSs), P-2OC and P-3OC, for spectral image compression and reconstruction. For this type of ICS, the weighting table or compression matrix H is modelled as a variable. The associated reconstruction matrix N is chosen as the Weiner estimation matrix. The objective function f(H) is the combination of the averages of the CIEDE2000 color difference (ΔE₀₀) and the root mean square error (RMSE) between the original and reconstructed reflectance values. Hence, the compression matrix H is determined by solving the nonlinear minimization problem based on the Munsell training dataset. The proposed ICSs, P-2OC and P-3OC, were tested and compared, respectively, with ICS-2SI and ICS-3SI developed by Zhang et al. (JOSAA, Vol. 29, pp. 1027–1034) in 2012 using the NCS dataset, and 2 spectral images. Performance tests showed that the proposed P-2OC and P-3OC ICSs are better than the ICS-2SI and ICS-3SI ICSs, respectively, in terms of RMSE, goodness of fit coefficient (GFC), and ΔE₀₀ under CIE illuminants D65, A, C and F11. Therefore, it is expected that the P-2OC and P-3OC ICSs can find applications in spectral image compression and cross-media reproduction.

Due to the limitation of model coloring process, 3D solid sand table cannot satisfactorily restore the color information of electronic sand table, which has become a technical bottleneck for the development of solid sand table model to higher accuracy and color restoration. This paper mainly studies the coating treatment of three-dimensional spray painting surface; independently prepares the coating solution, coating the three-dimensional surface, analyzes the physical properties and printing properties of the coating; solves the problems of ink absorption, ink color rendering and dot expansion of three-dimensional spray painting coating; and improves the color reduction of three-dimensional spray painting. The specific research conclusions are as follows: with the decrease of pigment particle size, the whiteness of the coating first increases and then decreases, the contact angle of the coating gradually decreases, and the color density of the coating gradually increases. When VAE-707 and SBR are used as adhesives, the whiteness of the coating is improved; with the decrease of the ratio of VAE-705 to PVA-1788, the whiteness decreased first and then increased. When the ratio of VAE-705 to PVA-1788 was 2:8, the whiteness was higher. VAE-707 as adhesive has the highest color density, followed by PVA-1799; When the ratio of VAE-705 to PVA-1788 is 6:4, the color density of the printing coating is higher and the ability of tone reproduction is better.

Printed electronics is a new technology for manufacturing electronic products. However, the existing printing technologies could not meet the requirements for high precision of printing as required for producing silicon microelectronic devices. A regulation method is proposed to solve the problem of forming precision control of the pattern in gravure printing electronics. Based on the Navier–Stokes equations and Continuity equation for Newton flow, the relationship between ink flow velocity and scraper pressure change is studied. Thus, the coupling effect of cell ceiling, scraper shape and ink is analyzed by numerical decoupling method. ANSYS CFX is used to simulate the velocity of the ink transferring process. The scraper deformation and ink film thickness uniformity were tested and measured. According to the simulation and test results, the model is modified. Research findings: Under the printing pressure, the deformation in the middle of the doctor blade is slightly lower than that at both ends. When the initial printing pressure is constant, most ink uneven streaks appear at 1/2-1/3 of the doctor blade. It is shown that increasing the initial angle of the scraper can improve the printing precision of the gravure printing electronic equipment via the reasonable matching of the initial angle by the roller scraper pressure.