Search

The rise of comics and games has led to increased artistic processing of portrait photos. With growing commercial demand and advancements in deep learning, neural networks for rapid facial style transfer have become a key research area in computer vision. This involves converting face photos into different styles while preserving content. Face images are more complex than regular images, requiring extensive modification. However, current methods often face issues such as unnatural color transitions, loss of detail in highlighted areas, and noticeable artifacts along edges, resulting in low-quality stylized images. In this study, an enhanced generative adversarial network (GAN) is proposed, which is based on Adaptive Layer Instance Normalization (AdaLIN) + Laplacian. This approach incorporates the AdaLIN normalization method, allowing for the dynamic adjustment of Instance Normalization (IN) and Layer Normalization (LN) parameters' weights during training. By combining the strengths of both normalization techniques, the model selectively preserves and alters content information to some extent, aiming to strike a balance between style and content. This helps address problems such as unnatural color transitions and loss of details in highlights that lead to color inconsistencies. Furthermore, the introduction of a Laplacian regularization term aids in denoising the image, preventing noise features from interfering with the color transfer process. This regularization also helps reduce color artifacts along the face's edges caused by noise while maintaining the image's contour information. These enhancements significantly enhance the quality of the generated face images. To compare our method with traditional CycleGAN and recent algorithms such as XGAN and CariGAN, both subjective and objective evaluations were conducted. Subjectively, our method demonstrates more natural color transitions and superior artifact elimination, achieving higher scores in Mean Opinion Score (MOS) evaluations. Objectively, experiments using our method yielded better scores across three metrics: FID, SSIM, and MS-SSIM. The effectiveness of the proposed methods is validated through both objective and subjective evaluations. [ABSTRACT FROM AUTHOR]

The aim of this study was to prepare and characterize stable non−covalent ternary complexes based on pea protein (PP, 0.5%), hyaluronic acid (HA, 0.125%), and chlorogenic acid (CA, 0~0.03%). The ternary complexes were comprehensively evaluated for physicochemical attributes, stability, emulsifying capacities, antioxidant properties, and antimicrobial efficacy. PP-HA binary complexes were first prepared at pH 7, and then CA was bound to the binary complexes, as verified by fluorescence quenching. Molecular docking elucidated that PP interacted with HA and CA through hydrogen bonding, hydrophobic and electrostatic interactions. The particle size of ternary complexes initially decreased, then increased with CA concentration, peaking at 0.025%. Ternary complexes demonstrated good stability against UV light and thermal treatment. Emulsifying activity of complexes initially decreased and then increased, with a turning point of 0.025%, while emulsion stability continued to increase. Complexes exhibited potent scavenging ability against free radicals and iron ions, intensifying with higher CA concentrations. Ternary complexes effectively inhibited Staphylococcus aureus and Escherichia coli, with inhibition up to 0.025%, then decreasing with CA concentration. Our study indicated that the prepared ternary complexes at pH 7 were stable and possessed good functionality, including emulsifying properties, antioxidant activity, and antibacterial properties under certain concentrations of CA. These findings may provide valuable insights for the targeted design and application of protein-polysaccharide-polyphenol complexes in beverages and dairy products. [ABSTRACT FROM AUTHOR]

Zinc-doped CuAl2O4 (Zn-CuAl2O4) nanofiber membranes with a high specific surface area, multi-pore structure, and a partial inverse spinel structure were successfully fabricated by controlling the electrospinning process and calcination atmospheres. The excessive doping of Zn2+ ions results in the formation of ZnAl2O4/CuAl2O4 fiber membrane, which has a poor UV-Vis absorption ability. The diameter of Zn-CuAl2O4 nanofibers is about 80 nm, and the specific surface area of the Zn-CuAl2O4 nanofiber membrane is 45.2 m2/g. XRD and XPS results indicate that Zn-CuAl2O4 nanofiber membranes contain spinel and inverse spinel phases. Compared with pure spinel CuAl2O4 nanofiber membranes, the UV-Vis absorption ability of the Zn-CuAl2O4 nanofiber membrane is enhanced, and transient photocurrent intensity was increased from 0.14 mA/cm2 to 0.16 mA/cm2. The free radical capture experiments and EPR spectra indicate that huge ·O2− are generated on the surface of the Zn-CuAl2O4 nanofiber membrane, and the photocatalytic degradation efficiency of MB is increased by 14%. The Zn-CuAl2O4 nanofiber membranes exhibit excellent cycling stability and have great potential for photocatalytic degradation of dye wastewater. [ABSTRACT FROM AUTHOR]

In order to determine metal ions in aqueous solution, a millisecond pulsed solution cathode glow discharge (SCGD) source coupled with a portable spectrometer was developed for atomic emission spectrometry (AES). The millisecond pulsed SCGD source was operated with two high-voltage DC power supplies and a high-voltage pulse switch, specifically boosting the analyte excitation and sustain the glow discharge. The method to excite the plasma discharge contrasts with traditional approaches that employ a single power source. The influence of operating parameters, including pulsed voltage, pulsed frequency, duty ratio, and sample flow rate, were optimized to construct a system having good stability and improved excitation of metal ions. The optimum detection limits for Na, K, Li, Mg, Ag, Cs, Rb, and Sr were 0.17, 0.32, 0.38, 0.50, 2.65, 1.30, 0.22, and 74.02 µg·L−1, respectively. Furthermore, mountain spring, Yangtze River, and reservoir water were spiked with Ag, Rb, and Cs; the recoveries ranged from 105% to 106%, 106% to 107%, and 107% to 108% indicating that the developed system has favorable analytical performance. [ABSTRACT FROM AUTHOR]