ID 原文 译文
23385 仿真结果表明，当干扰子空间已知时，该文所提出的检测器可完全抑制干扰，当干扰子空间未知时，所提检测器可有效抑制位于信号子空间的正交补空间内的干扰。 Simulation results show that the proposed detectors can suppress the interference effectively when the interference subspace is known, and can suppress the interference lying in the orthogonal complement space of the nominal signal subspace when the interference is completely unknown.
23386 深度信息被证明是人类视觉的重要组成部分，然而大部分显著性检测工作侧重于 2 维图像上的方法，并不能很好地利用深度进行 RGB-D 图像显著性检测。该文提出一种融合显著深度特征的 RGB-D 图像显著目标检测方法，提取基于颜色和深度显著图的综合特征，根据构图先验和背景先验的方法进行显著目标检测。 Depth information is proved to be an important part of human vision. However, most saliency detection methods based on 2D images do not make good use of depth information, thus an effective saliency detection method for RGB-D image is presented. It extracts color feature combined with depth saliency feature and detects salient objects based on photographic composition prior and background prior.
23387 首先，对原始深度图进行预处理：使用背景顶点区域、构图交点和紧密度处理深度图，多角度融合形成深度显著图，并作为显著深度特征，结合颜色特征形成综合特征； First, original depth map is preprocessed to form depth saliency feature by background vertex area, photographic composition intersections, and compactness method. Then the association matrix is constructed by the adjacency weight of comprehensive feature.
23388 其次，从前景角度，将综合特征通过边连接权重构造关联矩阵，根据构图先验，假设多层中心矩形为前景种子，通过流形排序方法计算出 RGB-D 图像的前景显著图； Manifold ranking is running from foreground view to form foreground saliency map based on photographic composition prior and fusion of depth saliency feature and color feature.
23389 从背景角度，根据背景先验以及边界连通性计算出背景显著图； In order to correct the error caused by assumption, the boundary connectivity is used to suppress background from background view.
23390 最后，将前景显著图和背景显著图进行融合并优化得到最终显著图。 Final saliency map builds on fusion of foreground and background saliency map.
23391 实验采用 RGB-D1000 数据集进行显著性检测，并与 4 种不同的方法进行对比，所提方法的显著性检测结果更接近人工标定结果，PR(查准率-查全率)曲线显示在相同召回率下准确率高于其他方法。 Experiments compared with 4 different methods on RGB-D1000 database show that the proposed method has better precision-recall curve and outperforms the state- of-the-art methods.
23392 为解决传统均匀 FFT 波束形成算法引起的 3 维声呐成像分辨率降低的问题，该文提出分区域 FFT 波束形成算法。 In order to solve the problem that traditional uniform FFT beamforming algorithm reduces the resolution of 3-D sonar imaging, this paper presents a sub-region FFT beamforming algorithm.
23393 远场条件下，以保证成像分辨率为约束条件，以划分数量最少为目标，采用遗传算法作为优化手段将成像区域划分为多个区域。 In the far field, the imaging area is partitioned into multiple regions using genetic algorithm as the optimization method. The objective of the optimization process is to minimize the number of partitions, with the constraints being the imaging resolution.
23394 在每个区域内选取一个波束方向，获得每一个接收阵元收到该方向回波时的解调输出，以此为原始数据在该区域内进行传统均匀 FFT 波束形成。 In each region, a beam direction is selected to obtain the demodulated output when each receiving element receives the directional echo as the original data, and the traditional uniform FFT beamforming is performed in the region.