Cropping and attention based approach for masked face recognition.

The global epidemic of COVID-19 makes people realize that wearing a mask is one of the most effective ways to protect ourselves from virus infections, which poses serious challenges for the existing face recognition system. To tackle the difficulties, a new method for masked face recognition is proposed by integrating a cropping-based approach with the Convolutional Block Attention Module (CBAM). The optimal cropping is explored for each case, while the CBAM module is adopted to focus on the regions around eyes. Two special application scenarios, using faces without mask for training to recognize masked faces, and using masked faces for training to recognize faces without mask, have also been studied. Comprehensive experiments on SMFRD, CISIA-Webface, AR and Extend Yela B datasets show that the proposed approach can significantly improve the performance of masked face recognition compared with other state-of-the-art approaches.

The Effect of Key Electronic States on Excess Lithium Intercalation in Li2RuyMn1-yO3.

Excess reversible lithium storage is an alternative crucial strategy besides the expansion of redox centers to boost the capacity of layered cathodes. However, the mechanism of excess Li+ intercalation is far from being comprehended, indisputably hindering the development of layered cathodes. Herein, the comparative study of Li2RuyMn1-yO3 and Li2RuyTi1-yO3 by X-ray absorption and photoemission spectroscopies attempts to illustrate the origin. The charge transfer from Ru to Mn through TM-O π bonding interaction with the formation of O holes has been revealed in Li2RuyMn1-yO3, which originates from the inductive effect and the approaching energy level of Mn and Ru bands. The electronic state is thought to reduce the Coulomb repulsion of Li+ with the matrix, promoting excess Li+ intercalation. The results are instructive to the rational design of layered cathodes to achieve a larger reversible capacity in a wide voltage window.

An analytical solution for two-dimensional vacuum preloading combined with electro-osmosis consolidation using EKG electrodes.

China is a country with vast territory, but economic development and population growth have reduced the usable land resources in recent years. Therefore, reclamation by pumping and filling is carried out in eastern coastal regions of China in order to meet the needs of urbanization. However, large areas of reclaimed land need rapid drainage consolidation treatment. Based on past researches on how to improve the treatment efficiency of soft clay using vacuum preloading combined with electro-osmosis, a two-dimensional drainage plane model was proposed according to the Terzaghi and Esrig consolidation theory. However, the analytical solution using two-dimensional plane model was never involved. Current analytical solutions can't have a thorough theoretical analysis of practical engineering and give relevant guidance. Considering the smearing effect and the rectangle arrangement pattern, an analytical solution is derived to describe the behavior of pore-water and the consolidation process by using EKG (electro-kinetic geo synthetics) materials. The functions of EKG materials include drainage, electric conduction and corrosion resistance. Comparison with test results is carried out to verify the analytical solution. It is found that the measured value is larger than the applied vacuum degree because of the stacking effect of the vacuum preloading and electro-osmosis. The trends of the mean measured value and the mean analytical value processes are comparable. Therefore, the consolidation model can accurately assess the change in pore-water pressure and the consolidation process during vacuum preloading combined with electro-osmosis.

Copper metal-organic polyhedra nanorods with high intrinsic peroxidase-like activity at physiological pH for bio-sensing.

Enzyme-mimicking nanomaterials (nanozymes) have been used to catalyze enzyme-like reactions for various applications. Despite these favorable properties, peroxidase-mimicking nanozymes suffer from the optimum reaction occurring in acidic solutions, which greatly limits their applications in biological systems. Herein, we demonstrate that copper Metal-Organic Polyhedra (Cu-MOP) nanorods exhibit high peroxidase-mimicking activity at a physiological pH value. Cu-MOP nanorods catalyze the generation of strong chemiluminescence signals in the presence of luminol/H2O2 and the oxidation of 2,2'-azinobis-(3-ethylbenzthiazoline-6-sulphonate acid) (ABTS) by H2O2 produces a green color reaction in a neutral pH solution. The Cu-MOP nanorod driven generation of color reactions is applied to develop a H2O2 sensor, and furthermore is implemented to develop a glucose detection platform.

Cathodic-controlled and near-infrared organic upconverter for local blood vessels mapping.

Organic materials are used in novel optoelectronic devices because of the ease and high compatibility of their fabrication processes. Here, we demonstrate a low-driving-voltage cathodic-controlled organic upconverter with a mapping application that converts near-infrared images to produce images of visible blood vessels. The proposed upconverter has a multilayer structure consisting of a photosensitive charge-generation layer (CGL) and a phosphorescent organic light-emitting diode (OLED) for producing clear images with a high resolution of 600 dots per inch. In this study, temperature-dependent electrical characterization was performed to analyze the interfacial modification of the cathodic-controlled upconverter. The result shows that the upconverter demonstrated a high conversion efficiency of 3.46% because of reduction in the injection barrier height at the interface between the CGL and the OLED.