ID 原文 译文
23535 因此，开展FMCW SAR 的信号处理研究具有重要的现实意义和军事价值。 Therefore, it is significant to make a deep research of FMCW SAR.
23536 目前，国内外关于 FMCW SAR 的研究主要集中在其成像处理方法上，在回波模拟方向研究甚少，特别是对于既包含分布式运动目标又有静止目标下的真实场景下的 FMCW SAR 回波快速模拟方法更是鲜有涉及。 However, much attention is paid on the aspect of data focusing while little research concentrates on the raw data simulation, especially for moving targets.
23537 基于此，该文在 FMCW SAR 平台脉内连续运动的基础上推导了一种适用于动静目标的精确的 FMCW SAR 2 维频域模型，并基于该模型提出一种高效精确的 FMCW SAR 动静目标混合场景回波模拟的方法。 In view of these, this paper firstly derived an accurate 2-D spectrum model in the consideration of the in-pulses motion of FMCW SAR, and then a highly-efficient raw data simulation method for FMCW SAR with moving targets is proposed.
23538 点目标和场景目标的仿真试验验证了该模拟方法的有效性，效率分析发现该种模拟方法的效率远高于传统的仿真方法。 Point targets and real scene raw data simulation experiments are carried out to validate it and the its efficiencyis analyzed. Results show its high efficiency comparing with the conventional methods.
23539 将机器学习运用到视网膜血管分割当中已成为一种趋势，然而选取什么特征作为血管与非血管的特征仍为众所思考的问题。 How to apply machine learning to retinal vessel segmentation effectively has become a trend, however, choosing what kind of features for the blood vessels is still a problem.
23540 该文利用将血管像素与非血管像素看作二分类的原理，提出一种混合的 5D 特征作为血管像素与非血管像素的表达，从而能够简单快速地将视网膜血管从背景中分割开来。 In this paper, the blood vessels of pixels are regarded as a theory of binary classification, and a hybrid 5D features for each pixel is put forward to extract retinal blood vessels from the background simplely and quickly.
23541 其中 5D 特征向量包括 CLAHE (Contrast Limited Adaptive Histgram Equalization)，高斯匹配滤波，Hesse 矩阵变换，形态学底帽变换，B-COSFIRE(Bar-selective Combination Of Shifted FIlter REsponses)，通过将融合特征输入 SVM(支持向量机)分类器训练得到所需的模型。 The 5D eigenvector includes Contrast Limited Adaptive Histgram Equalization (CLAHE), Gaussian matched filter, Hessian matrix transform, morphological bottom hat transform and Bar-selective Combination Of Shifted Filter Responses (B-COSFIRE). Then the fusion features are input into the Support Vector Machine (SVM) classifier to train a model that is needed. The proposed method is evaluated on two publicly available datasets of DRIVE and STARE, respectively.
23542 通过在 DRIVE 和 STARE 数据库进行实验分析，利用 Se, Sp, Acc, Ppv, Npv, F1-measure 等常规评价指标来检测分割效果，其中平均准确率分别达到 0.9573 和 0.9575，结果显示该融合方法比单独使用 B-COSFIRE 或者其他目前所提出的融合特征方法更准确有效。 Se, Sp, Acc, Ppv, Npv, F1-measure are used to test the proposed method, and average classification accuracies are 0.9573 and 0.9575 on the DRIVE and STARE datasets, respectively. Performance results show that the fusion method also outperform the state-of-the-art method including B-COSFIRE and other currently proposed fusion features method.
23543 该文针对稀疏重构解相干问题，利用接收数据厅奇导值分解(SVD)后的大特征值对应的特征矢量，提出一种改进解相干方法。 To solve the problem of coherent sources using sparse reconstruction method, this paper proposes an improved method for solving coherent sources using the eigenvectors corresponding to the largest eigenvalues after Singular Value Decomposition (SVD) decomposition of received data.
23544 该方法通过迭代这一特征矢量来重构角度，无需知道信号源的数目，即可准确重构角度信息，实现解相干。 The method reconstructs the angle by iterating the feature vector, and reconstructs the angle information accurately without knowing the number of the signal source.