ID	原文	译文
4183	然后，通过引导滤波对透射率进行细化处理，再利用自适应容差机制对天空区域的透射率进行修正；	Then, the transmissivity was further refined by theguided filtering technique. After that, the transmissivity in sky area was revised by an adaptive tolerance mechanism.
4184	最后，将复原图像从 RGB 空间转换到 HSV 空间，对图像的亮度和对比度进行色彩补偿。	Finally,the restored image was converted from RGB space to HSV space, and thus the brightness and contrast of the images could becolor compensated.
4185	实验结果表明，所提算法复原图像具有较高的清晰度，在处理天空等明亮区域时也能取得较好的去雾效果。	Experimental results show that the proposed algorithm restores the images effectively and obtain prefer-able defogging results with regard to the processing of bright areas such as the sky areas.
4186	现有的 Fast-flux 域名检测方法在稳定性、针对性和流量普适性方面存在一些不足，为此提出一种基于DNS 流量的检测方法 Fast-flucos。	There are three weaknesses in previous Fast-flux domain name detection method on the aspects of stability,targeting, and applicability to common real-world DNS traffic environment. For this, a method based on DNS traffic,called Fast-flucos was proposed.
4187	首先，采用流量异常过滤和关联匹配算法，以提高检测的稳定性；	Firstly, the traffic anomaly filtering and association matching algorithms were used forimproving detection stability.
4188	然后，引入量化的地理广度、国家向量表和时间向量表特征，以加强对 Fast-flux 域名检测的针对性；	Secondly, the features, quantified geographical width, country list, and time list, were ap-plied for better targeting Fast-flux domains.
4189	最后，采用更合理的正负样本和包括深度学习在内的多种机器学习方法确定最佳分类器和最优特征组合，以尽量确保对真实 DNS流量的普适性。	Lastly, the feature extraction were finished by the more suitable samples fortrying to adapt to common real-world DNS traffic. Several machine learning algorithms including deep learning are triedfor determining the best classifier and feature combination.
4190	基于真实 DNS 流量的实验表明，Fast-flucos 的召回率、精确率和 ROC_AUC 分别达到了 0.998 6、0.976 7 和 0.992 9，均优于当前主流的 EXPOSURE、GRADE 和 AAGD 等检测方法。	The experimental result based on real-world DNS trafficshows that Fast-flucos' recall rate is 0.998 6, precision is 0.976 7, and ROC_AUC is 0.992 9, which are all better than thecurrent main stream approaches, such as EXPOSURE, GRADE and AAGD.
4191	针对目前因果一致性模型中存在的元数据传播开销大、操作时延、远程更新可见时延高等问题，提出了基于共享图和部分复制策略的分布式存储因果一致性模型。	In order to solve the problem of metadata propagation overhead, operation delay and remote update visibilitylatency in the current causal consistency model, a causal consistency model for distributed data stores based on the sharedgraph and partial replication strategy was proposed.
4192	该模型以共享图拓扑结构为基础，每个数据中心存放完整数据集的子集，同时，提出了共享稳定向量与混合逻辑时钟相结合的全局稳定策略，在保证因果关系的前提下，实现数据中心间的数据一致性。	This model was based on the topology of the shared graph, and eachdata center stored an arbitrary subset of the data. At the same time, the global stabilization strategy combining shared sta-ble vector and hybrid logical clocks was proposed to provide data consistency guarantees on the premise of ensuringcausality.