Although advanced single image deraining methods have been proposed, one main challenge remains: the available methods usually perform well on specific rain patterns but can hardly deal with scenarios with dramatically different rain densities, especially when the impacts of rain streaks and the veiling effect caused by rain accumulation are heavily coupled. To tackle this challenge, we propose a two-stage density-aware single image deraining method with gated multi-scale feature fusion. In the first stage, a realistic physics model closer to real rain scenes is leveraged for initial deraining, and a network branch is also trained for rain density estimation to guide the subsequent refinement. The second stage of model-independent refinement is realized using conditional Generative Adversarial Network (cGAN), aiming to eliminate artifacts and improve the restoration quality. In particular, dilated convolutions are applied to extract rain features at multiple scales and gated feature fusion is exploited to better aggregate multi-level contextual information in both stages. Extensive experiments have been conducted on representative synthetic rain datasets and real rain scenes. Quantitative and qualitative results demonstrate the superiority of our method in terms of effectiveness and generalization ability, which outperforms the state-of-the-art.

Although advanced single image deraining methods have been proposed, their generalization ability to real-world images is usually limited, especially when dealing with rain patterns of different densities, shapes, and directions. In order to improve the robustness and generalization of these deraining methods, we propose a novel density-aware single image deraining method with gated multi-scale feature fusion, which consists of two stages. In the first stage, a sophisticated physics model is leveraged for initial deraining and a network branch is utilized for rain density estimation to guide the subsequent refinement. The second stage of model-independent refinement is realized using conditional Generative Adversarial Network (cGAN), attempting to eliminate artifacts and improve the restoration quality. Extensive experiments have been conducted on the representative synthetic rain datasets and real rain scenes, demonstrating the superiority of our method in terms of effectiveness and generalization ability, which outperforms the state-of-the-arts.

On the basis of Time-Trigger Protocol (TTP), Time Triggered Ethernet (TTE) inherits the characteristics of the popularity, maturity and flexibility of traditional Ethernet technology, so as to support the comprehensive transmission of various services with different properties, high reliability and high performance in the same network. For the precise global clock synchronization to ensure the service quality of high priority services, TTE has gradually been widely used in aerospace, rail transit, automotive vehicle driving and other fields. Traffic scheduling is a key issue of TTE technology to meet the deterministic network performance. In this paper, network calculus theory is applied to the performance analysis of schedule partition and schedule porosity method with different ratios of Time-Triggered (TT) and Rate-Constraint (RC) traffics. We focus on the latency of RC frames under the condition that the TT traffic transmission requirements are firstly met. The simulation results justify the analysis validation. It is shown that neither of two methods is not always work well for ever, which it depends on the amount of TT traffics in the total traffics.