ID 原文 译文
16995 为提高分割的准确率，在MR图像中提取了416维影像组学特征并与128维通过卷积神经网络提取的高阶特征进行组合和特征约简，将特征约简后产生的298维特征向量用于分类学习。 In order to improve the accuracyof segmentation, 416 radiomics features extracted from multi-modal MR images and 128 CNN featuresextracted by a convolutional neural network are mixed. The feature vector consisting of 298 features forclassification learning are formed after a feature reduction process.
16996 为对算法的性能进行验证，在BraTS2017数据集上进行了实验，实验结果显示该文提出的方法能够快速检测并定位肿瘤，同时相比其它方法，整体分割精度也有明显提升。 In order to verify the performance of theproposed algorithm, experiments are carried out on the BraTS2017 dataset. The experimental results show thatthe proposed method can quickly detect and locate the tumor. The overall segmentation accuracy is improveddistinctly with respect to 4 state-of-the-art approaches.
16997 该文提出了一种仅依靠激光探测与测量数据，实现单视图遥感影像数字表面模型(DSM)重建的新方法。 A novel method for Digital Surface Model (DSM) reconstruction of single-view remote sensing imageis proposed which only relies on light detection and ranging data.
16998 该方法基于深度学习技术设计了一种编码-解码结构的语义分割网络，该网络采用多尺度残差融合的编码块与解码(MRFED)块从输入图像中提取语义信息，进而逐像素预测高度值； Based on deep learning technology, asemantic segmentation network with an encode-decode structure is designed.
16999 采用特征图跳跃级联的策略保留输入图像的细节特征和结构信息。 The network uses Multi-scaleResidual Fusion Encode and Decode (MRFED) blocks to extract semantic information from the input image,and then predicts the height value pixel by pixel, as well as adopts a strategy of skip connections with featuremaps to preserves the detailed features and structural information of the input image.
17000 该文采用了一个包含DSM数据的遥感影像公开数据集训练与测试模型， The model is trained and tested on a public dataset of remote sensing images containing DSM data.
17001 实验结果表明：DSM重建结果与真值的平均绝对误差(MAE)为2.1e-02，均方根误差(RMSE)为3.8e-02，结构相似性(SSIM)为92.89%，均优于经典的深度学习语义分割网络。 Experiments show that, the MeanAbsolute Error (MAE) between DSM reconstruction results and true values is 2.1e-02, the Root Mean SquareError (RMSE) is 3.8e-02, and the Structural SIMilarity (SSIM) is 92.89%, which are all better than the classicdeep learning semantic segmentation networks.
17002 实验证实该方法能够有效实现单视图遥感影像的DSM重建，具有较高的精度，以及较强的地物分布结构重建能力。 Experiments confirm that the method can effectivelyreconstruct the DSM of single-view remote sensing images with high accuracy, as well as the structure offeature distribution.
17003 云数据安全问题是制约云计算发展的重要因素之一。 The security of cloud data is one of the most important factors to obstruct the development of cloudcomputing.
17004 该文综述了云数据安全方面的研究进展，将云数据安全所涉及的云身份认证、云访问控制、云数据安全计算、虚拟化安全技术、云数据存储安全、云数据安全删除、云信息流控制、云数据安全审计、云数据隐私保护及云业务可持续性保障10方面相关研究工作纳入到物理资源层、虚拟组件层及云服务层所构成的云架构中进行总结和分析； Therefore, on the basis of proposed three-tiers cloud architecture that consists of physical resourceslayer, virtual component layer, and cloud service layer, this paper makes a detail survey on existing works thatfocus on the security of cloud data, which involves in cloud identify authentication, cloud access control, clouddata secure computing, virtualization, cloud data security storage, cloud data secure deletion, information flowcontrol, cloud data secure auditing, cloud data privacy preserving, and cloud business continuity, respectively.