To send content items to your account,
please confirm that you agree to abide by our usage policies.
If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account.
Find out more about sending content to .
To send content items to your Kindle, first ensure firstname.lastname@example.org
is added to your Approved Personal Document E-mail List under your Personal Document Settings
on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part
of your Kindle email address below.
Find out more about sending to your Kindle.
Note you can select to send to either the @free.kindle.com or @kindle.com variations.
‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi.
‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.
Road scene understanding is a critical component in an autonomous driving system. Although the deep learning-based road scene segmentation can achieve very high accuracy, its complexity is also very high for developing real-time applications. It is challenging to design a neural net with high accuracy and low computational complexity. To address this issue, we investigate the advantages and disadvantages of several popular convolutional neural network (CNN) architectures in terms of speed, storage, and segmentation accuracy. We start from the fully convolutional network with VGG, and then we study ResNet and DenseNet. Through detailed experiments, we pick up the favorable components from the existing architectures and at the end, we construct a light-weight network architecture based on the DenseNet. Our proposed network, called DSNet, demonstrates a real-time testing (inferencing) ability (on the popular GPU platform) and it maintains an accuracy comparable with most previous systems. We test our system on several datasets including the challenging Cityscapes dataset (resolution of 1024 × 512) with an Mean Intersection over Union (mIoU) of about 69.1% and runtime of 0.0147 s/image on a single GTX 1080Ti. We also design a more accurate model but at the price of a slower speed, which has an mIoU of about 72.6% on the CamVid dataset.
In this article a comparison study of the numerical methods for compressible two-phase flows is presented. Although many numerical methods have been developed in recent years to deal with the jump conditions at the fluid-fluid interfaces in compressible multiphase flows, there is a lack of a detailed comparison of these methods. With this regard, the transport five equation model, the modified ghost fluid method and the cut-cell method are investigated here as the typical methods in this field. A variety of numerical experiments are conducted to examine their performance in simulating inviscid compressible two-phase flows. Numerical experiments include Richtmyer-Meshkov instability, interaction between a shock and a rectangle SF6 bubble, Rayleigh collapse of a cylindrical gas bubble in water and shock-induced bubble collapse, involving fluids with small or large density difference. Based on the numerical results, the performance of the method is assessed by the convergence order of the method with respect to interface position, mass conservation, interface resolution and computational efficiency.
A 100-J-level Nd:glass laser system in nanosecond-scale pulse width has been constructed to perform as a standard source of high-fluence-laser science experiments. The laser system, operating with typical pulse durations of 3–5 ns and beam diameter 60 mm, employs a sequence of successive rod amplifiers to achieve 100-J-level energy at 1053 nm at 3 ns. The frequency conversion can provide energy of 50-J level at 351 nm. In addition to the high stability of the energy output, the most valuable of the laser system is the high spatiotemporal beam quality of the output, which contains the uniform square pulse waveform, the uniform flat-top spatial fluence distribution and the uniform flat-top wavefront.
The Montreal Cognitive Assessment (MoCA) is used for screening mild cognitive impairment (MCI), and the Beijing version (MoCA-BJ) is widely used in China. We aimed to develop a computerized tool for MoCA-BJ (MoCA-CC).
MoCA-CC used person-machine interaction instead of patient-to-physician interaction; other aspects such as the scoring system did not differ from the original test. MoCA-CC, MoCA-BJ and routine neuropsychological tests were administered to 181 elderly participants (MCI = 96, normal controls [NC] = 85).
A total of 176 (97.24%) participants were evaluated successfully by MoCA-CC. Cronbach's α for MoCA-CC was 0.72. The test–retest reliability (retesting after six weeks) was good (intraclass correlation coefficient = 0.82; P < 0.001). Significant differences were observed in total scores (t = 9.38, P < 0.001) and individual item scores (t = 2.18–8.62, P < 0.05) between the NC and MCI groups, except for the score for “Naming” (t = 0.24, P = 0.81). The MoCA-CC total scores were highly correlated with the MoCA-BJ total scores (r = 0.93, P < 0.001) in the MCI participants. The area under receiver–operator curve for the prediction of MCI was 0.97 (95% confidence interval = 0.95–1.00). At the optimal cut-off score of 25/26, MoCA-CC demonstrated 95.8% sensitivity and 87.1% specificity.
The MoCA-CC tool developed here has several advantages over the paper-pencil method and is reliable for screening MCI in elderly Chinese individuals, especially in the primary clinical setting. It needs to be validated in other diverse and larger populations.
The hollow mesoporous spheres (HMS) with cubic pore network have been synthesized via a simple two-step method. Two drugs of different molecules size, Aspirin and Gentamicin, were tested by one simple adsorption process. Up to 336 mg Aspirin molecules can be stored in 1.0 g HMS, while Gentamicin molecules of much larger size are much more difficult to be introduced into the pore channels of HMS. The same results can be obtained by using MCM-48 and MCM-41 as comparative mesoporous carriers. The HMS shows significantly higher storage amount of Aspirin than conventional MCM-48 and MCM-41 due to its hollow core structure. The release process of HMS-Aspirin, MCM-48-Aspirin and MCM-41-Aspirin are found to have a sustained-release property and follow a Fickian diffusion mechanism. Moreover, the HMS is suitable for storage of drug molecules of much smaller size.
A facile, surfactant-assisted, hydrothermal approach has been developed to synthesize lanthanide phosphate single-crystalline nanowires/nanorods with smooth surface, uniform diameter, and good crystallinity. The surfactant Pluronic P123 was found to play a crucial role on the uniform morphology of lanthanide phosphate single-crystalline nanowires/nanorods. Photoluminescence spectra of the lanthanide phosphate single-crystalline nanowires/nanorods show that these nanowires/nanorods have strong photoluminescent emissions in the ultraviolet-visible and near-infrared regions. The present work is a preliminary and significant step toward the potential luminescent and catalytic applications of lanthanide compound based one-dimensional nanostructures.
The present study was designed to study the mechanisms by which dietary conjugated linoleic acids (CLA) decrease serum cholesterol. Hamsters were fed a semi-synthetic diet containing 1 g cholesterol/kg diet with or without supplementation with 20 g linoleic acid (LA) and 20 g CLA/kg diet. After 8 weeks, serum fasting total cholesterol (TC) and triacylglycerol (TG) were significantly lower in the LA-supplemented and CLA-supplemented groups compared with those of the control (CTL) hamsters. In contrast to LA, CLA significantly lowered hepatic cholesterol but it increased the level of adipose tissue cholesterol, suggesting that the hypocholesterolaemic mechanism of CLA is different from that of LA. CLA decreased the activity of intestinal acyl CoA:cholesterol acyltransferase (ACAT) whereas LA had no effect on this enzyme. Consequently, CLA supplementation increased the faecal excretion of total neutral sterols, but it had no or little effect on the faecal acidic sterols. If the ACAT is associated with cholesterol absorption, the part of mechanisms by which CLA decreases serum cholesterol may involve down-regulation of intestinal ACAT activity.
Interfacial reactions of Co/Si0.76Ge0.24 and Co(Si0.76Ge0.24)/Si0.76Ge0.24 by pulsed KrF laser annealing as a function of energy density and pulse number were studied. For the Co/Si0.76Ge0.24 samples annealed at an energy density of 0.2–0.6 J/cm2, three germanosilicide layers, i.e., amorphous structure and/or nanocrystal, Co(Si1−xGex), and Co(Si1−xGex)2, were successively formed along the film-depth direction. At 0.3 J/cm2 Ge segregated to the underlying Si0.76Ge0.24 film, inducing strain relaxation in the residual Si0.76Ge0.24 film. At 0.8 J/cm2 the reacted region was mostly transformed to a single layer of Co(Si1−xGex)2, whereas Ge further diffused to the Si substrate. At 1.0 J/cm2, constitutional supercooling appeared. Even the Co(Si0.76Ge0.24) film used as the starting material for laser annealing could not prevent the occurrence of constitutional supercooling at energy densities >1.6 J/cm2. The energy densities at which Co(Si1−xGex) transformation to Co(Si1−xGex)2, Ge segregation to the underlying Si, and constitutional supercooling occurred were higher for the Co(Si0.76Ge0.24)/ Si0.76Ge0.24 system than for the Co/Si0.76Ge0.24 system. Higher energy density and/or pulse number enhanced the growth of Co(Si1−xGex)2 film. In the present study, the Co/Si0.76Ge0.24 samples subjected to annealing at 0.2 J/cm2 for 20 pulses produced a smooth Co(Si0.76Ge0.24)2 film without inducing Ge segregation out of the germanosilicide and strain relaxation in the unreacted Si0.76Ge0.24 film.
Sputtered A1N films developed for piezoelectric resonators are extremely chemically reactive. As-sputtered films react with boiling water resulting in a complete loss of the AIN bond structure. Experiments to determine the effect on chemical stability of annealing the sputtered films at 1000 °C, indicate annealing in an oxidizing gas leads to partial oxidation of AlN. Annealing in an inert gas prevents oxidation but does not protect the films from attack by boiling water. Annealing in a reducing gas followed by annealing in an inert gas renders A1N films stable in boiling water. A1N film structure and composition have been studied via Refractive Index, XRD, SIMS, SEM, AES, XPS and FTIR evaluations.
Email your librarian or administrator to recommend adding this to your organisation's collection.