Color Transfer Algorithm between Images Based on a Two-Stage Convolutional Neural Network.

A color transfer algorithm between images based on two-stage convolutional neural network (CNN) is proposed. The first stage network is based on VGG19 architecture as the backbone. The reference image-based color transfer (RICT) model was used to extract the features of the reference image and the target image, so as to realize the color transfer between them. The second stage is based on progressive convolutional neural network (PCNN) as its backbone. The palette-based emotional color enhancement (PECE) model is adopted to enhance the emotional coloring of the resulting image by comparing the palette, emotional value and the proportion of each color of the reference image. Through five sets of experiments, it is proved that the visual effect processed by our model is obviously better than several main colorization methods in both subjective evaluation and objective data. It can be applied to various complex scenes, and in the near future, it can also be better applied to the fields of digital restoration of old image archives, medical image coloring, art restoration, remote sensing image enhancement, infrared image enhancement and other fields.

Effect of Silicon Source (Fly Ash, Silica Dust, Gangue) on the Preparation of Porous Mullite Ceramics from Aluminum Dross.

Aluminum dross (AD) is a waste product produced during aluminum processing and can be used to prepare mullite ceramic materials. However, the research on the preparation of mullite porous ceramics entirely from solid waste is still in the development stage. In this paper, porous mullite ceramics were successfully fabricated using a solid-phase sintering process with AD and different silicon sources (fly ash, silica dust, and gangue) as raw materials. The bulk density, apparent porosity, and compressive strength of the specimens were obtained, and the phase compositions and microstructures of the sintered specimens were measured using XRD and SEM, respectively. The average activation energy of the phase transition of fly ash, silica dust, and gangue as silicon sources were 984 kJ/mol, 1113 kJ/mol, and 741 kJ/mol, respectively. The microstructures of the mullite in the specimens were prisms, random aggregates, and needle-shaped, respectively. The formation of needle-shaped mullite combined with the substrate enhanced the mechanical strength of the porous mullite ceramics. The apparent porosity, density, and compressive strength of the specimens with gangue as the silicon source were 33.13%, 1.98 g/cm3, and 147.84 MPa, respectively, when sintered at 1300 °C for 2 h.

Hydrolysis Behavior and Kinetics of AlN in Aluminum Dross during the Hydrometallurgical Process.

In this study, the hydrolysis behavior and kinetics of AlN in aluminum dross (AD) were investigated in order to better identify the steps controlling the AlN hydrolysis reaction and the factors influencing the hydrolysis rate to enhance the removal efficiency of AlN. The hydrolysis behavior of AlN, including AlN content, phase composition, chemical composition, microstructure, and element distribution, was determined by a leaching test, X-ray diffraction, X-ray fluorescence, scanning electron microscopy, and energy dispersive spectroscopy, respectively. The results showed that increasing the leaching liquid-solid ratio as well as the temperature was helpful for the removal efficiency of AlN. When the liquid--solid ratio was 4:1, temperature was 90 °C, and leaching time was 300 min, the removal efficiency of AlN reached 89.05%. The kinetics were described using the unreacted core model, and when the temperature was 30-40, 50-70, and 80-90 °C, the hydrolysis reaction rate of AlN was controlled by boundary layer diffusion, chemical reaction control, and product layer diffusion control, respectively.

Waste to Wealth Strategy: Preparation and Properties of Lightweight Al2O3-SiO2-Rich Castables Using Aluminum Dross Waste.

Aluminum dross is a well-known industrial waste generated in the aluminium industry, and its recycling and reuse is still a worldwide issue. Herein, aluminum dross waste (ADW) was recycled to progressively replace the aggregate fraction of clay at 70, 75, 80, 85, and 90 wt% for the fabrication of Al2O3-SiO2-rich porous castable refractories. Their physical properties and mechanical behavior were assessed by the measurement of linear shrinkage rate, bulk density, apparent porosity, cold crushing strength, and thermal conductivity. The microstructure and phase evolutions were analyzed via scanning electron microscopy (SEM) and X-ray diffraction (XRD). The incorporation of 85 wt% of ADW allowed the development of a waste-containing conventional refractory castable with improved properties as compared to those of the other samples. The sustainable refractory castable exhibited decent thermal conductivity and physical and mechanical characteristics, and is suitable for application as reheating furnace lining. It is a "green" practice to partially replace the traditional raw materials with industrial waste in the manufacture of conventional refractory castables and provides environmental and economic benefits.

Fully automatic image colorization based on semantic segmentation technology.

Aiming at these problems of image colorization algorithms based on deep learning, such as color bleeding and insufficient color, this paper converts the study of image colorization to the optimization of image semantic segmentation, and proposes a fully automatic image colorization model based on semantic segmentation technology. Firstly, we use the encoder as the local feature extraction network and use VGG-16 as the global feature extraction network. These two parts do not interfere with each other, but they share the low-level feature. Then, the first fusion module is constructed to merge local features and global features, and the fusion results are input into semantic segmentation network and color prediction network respectively. Finally, the color prediction network obtains the semantic segmentation information of the image through the second fusion module, and predicts the chrominance of the image based on it. Through several sets of experiments, it is proved that the performance of our model becomes stronger and stronger under the nourishment of the data. Even in some complex scenes, our model can predict reasonable colors and color correctly, and the output effect is very real and natural.

Thermo-Mechanical Coupling Analyses for Al Alloy Brake Discs with Al2O3-SiC(3D)/Al Alloy Composite Wear-Resisting Surface Layer for High-Speed Trains.

In the present work, a theoretical model of three-dimensional (3D) transient temperature field for Al alloy brake discs with Al2O3-SiC(3D)/Al alloy wear-resisting surface layer was established. 3D transient thermo-stress coupling finite element (FE) and computational fluid dynamic (CFD) models of the brake discs was presented. The variation regularities of transient temperature and internal temperature gradient of the brake discs under different emergency braking conditions were obtained. The effects of initial braking velocity (IBV) and thickness of Al2O3-SiC(3D)/Al alloy composite wear-resisting layer on the maximum friction temperature evolution of the disc were discussed. The results indicated the lower temperature and thermal stress distributed uniformly on the wear-resisting surface, which was dominated by high conductivity and cooling ability of the Al alloy brake disc. The maximum friction temperature was not obviously affected by the thickness of the wear-resisting layer. The maximum friction temperature of the brake discs increased with the increase of the IBV, the maximum friction temperature and thermal stress of the brake discs is about 517 °C and 192 MPa at IBV = 97 m/s considering air cooling, respectively. The lower thermal stress and fewer thermal cracks are produced during the braking process, which relatively decrease the damage. The friction behavior of the tribo-couple predicted using FE method correlated well with the experimental results obtained by sub-scale testing.