Search

Surface defect detection is a crucial step in ensuring the quality of lenses. One method to check for surface defects is to use an optical system integrated with an industrial camera to magnify and highlight the position of a defect on the surface of a lens. Therefore, automatic optical inspection systems are applied to detect micro-defects. In this study, we propose an automatic inspection platform based on a deep neural network for automatically imaging and examining the surface of a lens. High-resolution images of 2448 × 2048 pixels are acquired using a hybrid lighting system. A convolutional neural network integrated with a trainable Gabor filter is used as a machine vision algorithm to perform image classification and defect segmentation tasks. The experimental results show that the proposed method effectively performed with noise in the background, achieving a segmentation accuracy of 98%.

With the rapid development of display technology, related diseases of the human eye are also increasing day by day. Eye floaters are one of the diseases that affect humans. Herein, we present a functional ophthalmic dressing that can permeate the skin tissues of the eyes through oxygen and hydrogen to improve the symptoms of floaters. In clinical tests, the symptoms of sensory floaters improved in 28 patients, and the recovery rates of mild, moderate, and severe floaters were about 70%, 66.7%, and 83.3%, respectively.

The file attached to this record is the author's final peer reviewed version. The Publisher's final version can be found by following the DOI link. We examine the effects of CEO compensation, excess reserves, and role of monetary policy on bank risk-taking behaviour based on a sample of 88 Chinese commercial banks over the period of 2003-2014. We find evidence that suggests that incentives present in CEO compensation contracts and excess reserves exert a positive and significant impact on risk-taking and credit risk. However, we find that the positive effects of CEO compensation and excess reserves on risk-taking are cancelled out by the interaction of CEO compensation and excess reserves. Further analysis suggests that the central bank’s monetary policy serves to restrain the effects of an interaction between CEO compensation and excess reserves on bank risk-taking and credit risk. This study extends the theoretical model, which indicates that excess reserves are a major source of credit risk and notes that the effects of incentives inherent in CEO compensation contracts and excess reserves on bank risk policies are contingent on the monetary policy pursued by the central bank in China’s emerging economy

As an important task of the country's socio-economic development, protecting the rights offemale workers in employment and income is concerned and implemented by international organizationsand all countries in the world. The COVID-19 pandemic has been affecting the global economy and hurtthe incomes and employment of many female workers. In Vietnam nowadays, there is an increase in theunemployment rate of female workers, gender inequality in employment and income, and the ability tosecure their jobs and income. The article analyzes the current regulations and their implementation in theemployment and income of Vietnamese female workers in the context of the COVID-19 pandemic. Thearticle also proposes some recommendations on legal provisions related to training backup jobs, arrangingand employing female employees, providing income support for female employees during leave due tothe COVID-19 epidemic; and policies related to female workers and their employers. Theserecommendations will improve Vietnam's labor law on female workers’ rights in employment and incomeand enhance the efficiency of human resource use and socio-economic development.

The left hemiface expresses emotion more intensely than the right. Because emotional expressions contract the facial muscles and wrinkle the skin, theoretically the left hemiface’s greater expressivity should prompt more pronounced expression lines and wrinkles on the left than right side of the face. As wrinkles are the most salient age cue, we investigated whether the left hemiface consequently appears older than the right. Two hundred and sixty participants (F=148; M=112) viewed booklets containing pairs of left-left and right-right chimeric faces of eight models (M=F). For each trial participants were asked to make a two alternative forced choice response indicating which image looked older. Results confirmed a left cheek bias, with participants more likely to select left-left than right-right chimeras. Whilst participant gender did not influence perceptions, model gender predicted cheek selections: responses to female models drive the overall left cheek bias. The left cheek (56.8%) appeared older than the right cheek (43.2%) for female models, whereas there was little difference in perceived age between male models' left (50.8%) and right (49.2%) cheeks. Given that youth influences perceptions of female beauty, these findings complement previous research and offer a potential explanation for why the left side of females’ faces are judged less attractive: the right cheek appears younger.

This paper empirically investigates the three-way linkages amongst foreign direct investment (FDI), shadow economy and institutional quality by applying the panel dynamic simultaneous-equation modelling approach for a sample of 19 developing Asian countries over the period of 2002–2015. The empirical results by two-step System GMM show that institutional quality attracts inward FDI and FDI in its turn improves institutional quality, institutional quality is not only the cause but also the consequence of the shadow economy, and FDI inflows help reduce shadow economies though the channel of institutional improvement and lower shadow economies – which increase institutional quality – encourage FDI inflows. The empirical insights suggest helpful policy implications to deal with these dynamics simultaneously.

A methodology that applies hyperspectral imaging (HSI) on ophthalmoscope images to identify diabetic retinopathy (DR) stage is demonstrated. First, an algorithm for HSI image analysis is applied to the average reflectance spectra of simulated arteries and veins in ophthalmoscope images. Second, the average simulated spectra are categorized by using a principal component analysis (PCA) score plot. Third, Beer&ndash, Lambert law is applied to calculate vessel oxygen saturation in the ophthalmoscope images, and oxygenation maps are obtained. The average reflectance spectra and PCA results indicate that average reflectance changes with the deterioration of DR. The G-channel gradually decreases because of vascular disease, whereas the R-channel gradually increases with oxygen saturation in the vessels. As DR deteriorates, the oxygen utilization of retinal tissues gradually decreases, and thus oxygen saturation in the veins gradually increases. The sensitivity of diagnosis is based on the severity of retinopathy due to diabetes. Normal, background DR (BDR), pre-proliferative DR (PPDR), and proliferative DR (PDR) are arranged in order of 90.00%, 81.13%, 87.75%, and 93.75%, respectively, the accuracy is 90%, 86%, 86%, 90%, respectively. The F1-scores are 90% (Normal), 83.49% (BDR), 86.86% (PPDR), and 91.83% (PDR), and the accuracy rates are 95%, 91.5%, 93.5%, and 96%, respectively.

Light-emitting diodes (LEDs) have numerous advantages. However, LEDs only offer a point light source. Therefore, transforming LEDs into planar light sources is a new objective in general lighting applications. Solid light guides have strong uniformity but are marred by their material absorption characteristics and weight. Hollow light guides constitute a solution to the weight problem but exhibit poor uniformity and necessitate sacrificing efficiency to enhance uniformity. To resolve the uniformity, weight, and efficiency problems simultaneously, we propose a hollow light guide architecture involving mini-LEDs with asymmetric luminous intensity. To develop this guide module, we first optimized the aspect ratio of the cavity and then modulated the light path by using varied angles of the reflection surface on the end wall of the module. We then designed a beveled reflection surface near the mini-LEDs to further enhance uniformity. An archetype of the proposed architecture for planar light source modules had a width and depth of 51.5 and 9.95 mm, respectively. Experimental results revealed a total efficiency of 83.9% and uniformity of 92.3%. The module weight was determined to be 215 g, which was 40% lighter than that of similarly sized solid light guide modules.

InGaN/GaN samples grown on c-plane sapphire substrate with different In concentrations by metal organic chemical vapor deposition are demonstrated. The subsequent capping GaN layer growth opens a possibility for dislocation reduction due to the lateral strain relaxation in growth geometry. We present the further growth optimization and innovative characterization of InGaN layers overgrown on different structures with varying In concentrations. The photoelectrical and optical properties of the InGaN layers with or without capping GaN layer were investigated by time-resolved picosecond transient grating and temperature dependence photoluminescence. We note a 10-fold increase in carrier lifetime in the InGaN layers when the sample structure changes from PIN to single InGaN layer.

This paper proposes a new encapsulation structure for aluminum nitride-based deep UV light-emitting diodes (DUV-LEDs) and eutectic flip chips containing polydimethylsiloxane (PDMS) fluid doped with SiO2 nanoparticles (NPs) with a UV-transparent quartz hemispherical glass cover. Experimental results reveal that the proposed encapsulation structure has considerably higher light output power than the traditional one. The light extraction efficiency was increased by 66.49% when the forward current of the DUV-LED was 200 mA. Doping the PDMS fluid with SiO2 NPs resulted in higher light output power than that of undoped fluid. The maximum efficiency was achieved at a doping concentration of 0.2 wt%. The optical output power at 200 mA forward current of the encapsulation structure with NP doping of the fluid was 15% higher than that without NP doping. The optical output power of the proposed encapsulation structure was 81.49% higher than that of the traditional encapsulation structure. The enhanced light output power was due to light scattering caused by the SiO2 NPs and the increased average refractive index. The encapsulation temperature can be reduced by 4 °C at a driving current of 200 mA by using the proposed encapsulation structure.

An artificial intelligence algorithm to detect mycosis fungoides (MF), psoriasis (PSO), and atopic dermatitis (AD) is demonstrated. Results showed that 10 s was consumed by the single shot multibox detector (SSD) model to analyze 292 test images, among which 273 images were correctly detected. Verification of ground truth samples of this research come from pathological tissue slices and OCT analysis. The SSD diagnosis accuracy rate was 93%. The sensitivity values of the SSD model in diagnosing the skin lesions according to the symptoms of PSO, AD, MF, and normal were 96%, 80%, 94%, and 95%, and the corresponding precision were 96%, 86%, 98%, and 90%. The highest sensitivity rate was found in MF probably because of the spread of cancer cells in the skin and relatively large lesions of MF. Many differences were found in the accuracy between AD and the other diseases. The collected AD images were all in the elbow or arm and other joints, the area with AD was small, and the features were not obvious. Hence, the proposed SSD could be used to identify the four diseases by using skin image detection, but the diagnosis of AD was relatively poor.

Increasing attention has been paid to two-dimensional (2D) materials because of their superior performance and wafer-level synthesis methods. However, the large-area characterization, precision, intelligent automation, and high-efficiency detection of nanostructures for 2D materials have not yet reached an industrial level. Therefore, we use big data analysis and deep learning methods to develop a set of visible-light hyperspectral imaging technologies successfully for the automatic identification of few-layers MoS2. For the classification algorithm, we propose deep neural network, one-dimensional (1D) convolutional neural network, and three-dimensional (3D) convolutional neural network (3D-CNN) models to explore the correlation between the accuracy of model recognition and the optical characteristics of few-layers MoS2. The experimental results show that the 3D-CNN has better generalization capability than other classification models, and this model is applicable to the feature input of the spatial and spectral domains. Such a difference consists in previous versions of the present study without specific substrate, and images of different dynamic ranges on a section of the sample may be administered via the automatic shutter aperture. Therefore, adjusting the imaging quality under the same color contrast conditions is unnecessary, and the process of the conventional image is not used to achieve the maximum field of view recognition range of ~1.92 mm2. The image resolution can reach ~100 nm and the detection time is 3 min per one image.

The file attached to this record is the author's final peer reviewed version. The Publisher's final version can be found by following the DOI link. This paper reviews academic studies of excess reserves in the banking system of emerging economies from 2000 through 2014. While excess reserves in emerging countries have attracted increasing attention from scholars, virtually no work has reviewed and synthesised the extant knowledge. This paper takes the necessary step of consolidating and integrating the past literature on emerging country excess reserves. Focusing on articles published in major scholarly journals, we classify the existing literature on excess reserves into three broad taxonomies, namely excess liquidity sources, excess liquidity’s effects, and the response policies of central banks of emerging countries. Achievements within each of the three research areas are reviewed, critical gaps identified, and recommendations for future research provided.

Using a sample of 95 banks that covers the period 2000–2011, this article examines Chinese banks’ credit lending behaviour in response to the changes in the reserve requirement ratio in the presence of involuntary excess reserves (IERs) in the banking system. The study finds that Chinese banks with positive IERs one period after a reserve requirement shock experience a significantly increased credit supply in response to an increase in reserve requirement ratio. However, the reserve requirements have no significant impact on the credit supply in Chinese banks that have negative IERs one period after a reserve requirement shock. This article sheds lights on the effectiveness of Chinese monetary policy, which uses reserve requirements as the primary tool to sterilize excess liquidity and restrain credit expansion.

In this paper, we examine the effects of monetary policy on the risk-taking behaviour of Chinese banks in the presence of involuntary excess reserves based on a sample of 95 banks. We find that involuntary excess reserves lead to more aggressive risk-taking suggesting that large involuntary excess reserves stimulate the rapid expansion of credit and the price bubble in the Chinese financial market. However, banks with larger involuntary excess reserves tend to reduce risk-taking more rapidly under the tightening monetary policy regime. The paper sheds light on the effectiveness of government monetary policy in reducing the risk-taking behaviour of banks in an emerging market where involuntary excess reserves are present.

This study examines the bank lending channel and the impact of bank-specific characteristics on credit supply in response to monetary policy changes in respect of involuntary excess reserves in China. Using a sample of 95 banks, this study finds that the interaction of involuntary excess reserves with bank liquidity and with large bank size is more vulnerable to monetary policy shocks, while capitalisation has an insignificant impact on credit supply in response to monetary policy changes. Our results indicate that involuntary excess reserves, the size of the bank and liquidity are important in assessing the effect of monetary policy on loan growth. This implies that an agency problem, such as volume-based compensation, appears to drive bank lending and risk-taking behaviour in China.

We consider an elastic thin film as a bounded open subset ω of ℝ2. First, the effective energy functional for the thin film ω is obtained, by Г-convergence and 3D-2D dimension reduction techniques applied to the sequence of re-scaled total energy integral functionals of the elastic cylinders ω × ( - ε 2 , ε 2 ) $\omega \times \left( { - {\varepsilon \over 2},{\varepsilon \over 2}} \right)$ as the thickness ε goes to 0. Then we prove the existence of minimizers of the film energy functional. These results are proved in the case when the energy density function for the elastic cylinders has the growth prescribed by an Orlicz convex function M. Here M is assumed to be non-power-growth-type and to satisfy the conditions ∇2 and ∇2 (that is equivalent to the reflexivity of Orlicz and Orlicz–Sobolev spaces generated by M). These results extend results of H. Le Dret and A. Raoult for the case M(t) = |t| p for some p ∈ (1, ∞).

3D full head synthesis (3DFHS) is of interest in many applications, including gaming and social networks. One well known related work is [Blanz and Vetter 1999] but it is only restricted to the synthesis of the face area (and not the head area). FaceGen [FaceGen 2011] uses a similar statistical approach and it provided a solution by coloring the head area with a single matching color. However, both of these approaches are not automatic, and user interactions are needed to manually assign the locations of the feature points. In addition, computation time to perform 3D face synthesis is in the order of several minutes on relatively fast PCs.

This paper describes a system for automatic and real-time 3D photo-realistic face synthesis from a single frontal face image. This system employs a generic 3D head model approach for 3D face synthesis which can generate the 3D mapped face in real-time. The system first automatically detects face features from the input face image that corresponds to landmark points on the generic 3D head model. Thereafter, the generic head model is deformed to match the detected features. Finally, the texture from the face image is mapped onto the deformed 3D head model to create a photo-realistic 3D face. The system has the advantage of being totally automatic and runs in real-time. Experiments conducted show that good results can be obtained with no user intervention. Such a system is useful in many applications such as creation of avatar for virtual world by end-user with no need for manual and tedious process such as manual feature placements on the face image.