ID 原文 译文
20325 在实验测试集上3个指标分别达到了96.43%, 3.57%和96.25%，取得了令人满意的结果。 On the experimental test set,the above three indicators reach 96.43%, 3.57%, and 96.25% respectively, and satisfactory results are obtained.
20326 为提高现有随机脉冲噪声(RVIN)检测算法的检测准确率和执行效率，该文试图从构建描述能力更强的特征矢量和训练非线性映射更为准确的预测模型两个方面入手，实现一种基于训练策略的快速RVIN检测算法。 To improve the detection accuracy and execution efficiency of the existing Random-Valued Impulse Noise (RVIN) detectors, a fast training-based RVIN detection algorithm is implemented by constructing a more descriptive feature vector and training a detection model with more accurate nonlinear mapping.
20327 一方面，提取多个不同阶的对数绝对差值排序统计值并结合一个能够反映图像边缘特性的统计值作为刻画图块中心像素点是否为噪声的特征矢量。 On the one hand, multiple Rank-Ordered Logarithmic absolute Deviation (ROLD) statistics are extracted and combined with a statistical value reflecting the edge characteristics in the form of feature vector to describe how RVIN-like the center pixel of a patch is.
20328 在计算量增加极少的情况下，显著提升了特征矢量的描述能力。 The description ability of the feature vector is improved significantly while the computational complexity is just increased in small amount.
20329 另一方面，基于深度置信网络(DBN)训练RVIN预测模型(RVIN检测器)将特征矢量映射为噪声类型标签，实现了比浅层预测模型更为准确的映射。 On the other hand, an RVIN prediction model (RVIN detector) is obtained by training a Deep Belief Network (DBN) to map the feature vectors to noise labels, which is more accurate than the shallow prediction model.
20330 大量实验数据表明：与现有的RVIN检测算法相比，所提算法在检测准确率和执行效率两个方面都更有优势。 Extensive experimental results show that, compared with the existing RVIN detectors, the proposed one has better performance in terms of detection accuracy and execution efficiency.
20331 现有IP定位技术通过查询IP注册信息数据库或利用测量得到的时延等信息确定IP具体位置，在实际中由于受各种因素的影响，对网络中的大部分IP都无法得到准确、合理的定位结果。 The existing IP location technology determines the location of IP by querying IP to register information databases or using time-delay information. In fact, due to the influence of various factors, most ofthe IP in the network can not get accurate and reasonable positioning results.
20332 为此，该文提出一种基于网络结构特征的IP所属区域识别方法。 For this reason, a region recognition method of IP is proposed based on network structure features.
20333 该方法通过探测节点向待定位的IP发送Traceroute探测包获得两者之间的网络结构特征，并比较待定位节点和已知地理位置节点之间的网络结构特征确定待定位节点所属区域。 This method obtains the network topology information between the two nodes by sending the Trace route detection packet from the detection nodes to the IPs that need to be located Comparing the network structure features between the nodes to be located and the known geographical nodes determines where the nodes located.
20334 测试结果表明该文方法和现有的数据库查询的正确率相比有部分提升。 The actual test shows that thismethod can achieve better results.