A full-field transmission X-ray microscope (TXM) operating continuously from 5 keV to 12 keV with fluorescence mapping capability has been designed and constructed at the Beijing Synchrotron Radiation Facility, a first-generation synchrotron radiation facility operating at 2.5 GeV. Spatial resolution better than 30 nm has been demonstrated using a Siemens star pattern in both absorption mode and Zernike phase-contrast mode. A scanning-probe mode fluorescence mapping capability integrated with the TXM has been shown to provide 50 p.p.m. sensitivity for trace elements with a spatial resolution (limited by probing beam spot size) of 20 µm. The optics design, testing of spatial resolution and fluorescence sensitivity are presented here, including performance measurement results.

In this paper, we propose a cellular automata (CA) model for traffic flow in the framework of Kerner's three-phase traffic theory. We mainly consider the velocity-difference effect on the randomization of vehicles. The presented model is equivalent to a combination of two CA models, i.e., the Kerner-Klenov-Wolf (KKW) CA model and the Nagel-Schreckenberg (NS) CA model with slow-to-start effect. With a given probability, vehicle dynamical rules are changed over time randomly between the rules of the NS model and the rules of the KKW model. Due to the rules of the KKW model, the speed adaptation effect of three-phase traffic theory is automatically taken into account and our model can show synchronized flow. Due to the rules of the NS model, our model can show wide moving jams. The effect of "switching" from the rules of the KKW model to the rules of the NS model provides equivalent effects to the "acceleration noise" in the KKW model. Numerical simulations are performed for both periodic and open boundaries conditions. The results are consistent with the well-known results of the three-phase traffic theory published before.

No abstract

To promote the super-resolution (SR) technology in real-world applications, the blind SR, involving kernel estimation and image restoration to super-resolve images with unknown degradation, has become one of the research focuses. Most existing methods either implement the above two tasks step-by-step so that do not well consider the compatibility between them, or repeatedly apply two modules over and over again to emphasize cooperation but limit the adaptive development of each one. Towards the above issues, based on the Deep Alternating Network (DAN), a novel training strategy named switching the iteration is proposed in this paper. In the first stage, an estimation module and a restoration module are optimized alternately to promote compatibility. In the second stage, duplicate the pre-trained modules and place them alternately to form a linear structure to promote adaptive development. Extensive experiments on isotropic Gaussian degradation datasets and irregular blur kernel degradation datasets show that the proposed method can achieve visually pleasing results and state-of-the-art performance in blind SR.