ID 原文 译文
53647 最后设计了三重损失函数:模态间损失，模态内损失和汉明空间损失对模型进行训练学习。 Finally, a triple loss function is designed to train the model, including the inter-modal loss, the intra-modal loss, and hamming space loss.
53648 实验结果表明，本文所提方法在 MIＲFLICKＲ-25K 和 NUS-WIDE 数据集上均获得了良好的跨模态检索效果。 The experimental results on both MIＲFLICKＲ-25K and NUS-WIDE datasets show that the proposed method in this paper has obtained good cross-modal retrieval results.
53649 永久散射体法( Permanent Scatterer，PS) 是地基合成孔径雷达( Ground-Based Synthetic Aperture Ｒadar，GB- SAＲ) 形变监测的技术支撑， The selection of Permanent Scatterer ( PS) points is the key technology of Ground Based Synthetic Aperture Ｒa- dar ( GBSAＲ) deformation inversion.
53650 但使用传统多阈值法选取 PS 点时，会存在各个区域对阈值敏感性不同的问题。 However, when using the traditional threshold method to select PS points, there is a problem that each region has different sensitivity to the threshold.
53651 为解决选取 PS 点时漏选或错选的问题，本文提出一种注意力网络模型对 GBSAＲ 时序数据进行 PS 点筛选，并将该模型与循环神经网络( Ｒecurrent Neural Network，ＲNN) 和长短期记忆网络( Long Short-Term Memory，LSTM) 进行对比实验，监测三个不同场景来比较选取 PS 点的结果。 To solve the problem of missing or wrong selection when selecting PS points, this paper proposes a model based on attention network to process radar sequence data for PS point se- lection, compared with the recurrent neural network ( ＲNN) and long short term memory ( LSTM) by screening PS point of three regions.
53652 实验结果表明:基于注意力网络的模型实时性比 ＲNN 模型更好，准确度比 LSTM 模型更高。 The experimental results show that the real-time performance of the attention network based model is better than that of the ＲNN model, and its accuracy is higher than that of the LSTM model.
53653 因此基于注意力网络的模型在 PS 点选取上更具优势。 Therefore, the model based on atten- tion network has more advantages in PS point selection.
53654 针对复杂电磁环境中信号功率对入射信号波达方向( DOA) 估计的影响问题进行研究，发现用于 DOA 估计算法性能分析的经典评价准则对不同功率入射信号存在局限性。 In this paper, the effect of signal power on the DOA estimation of incident signals in complex electromagnetic environment is investigated. It is found that the classic evaluation criterion for performance analysis of DOA estimation algo- rithms has limitations on incident signals with different power.
53655 针对该问题，首先证明了强信号功率会影响弱信号 DOA 估计性能，得到强信号功率增加会导致弱信号功率克拉美罗界上升，即弱信号 DOA 估计的均方根误差增加。 Thus, this paper first demonstrates that the power of strong signal has influence on the DOA estimation performance of weak signal. The increasing of strong signal power will lead to the rising of Cramer-Ｒao Lower Bound of weak signal, that is, the increasing of the root mean square error.
53656 然后分析了 DOA 估计算法的经典评价准则对分辨不同功率入射信号存在的局限性，通过蒙特卡洛实验验证了经典评价准则对分辨不同功率入射信号存在较大误判率， Then this paper analyzes the limitations of the classic evaluation criterion for DOA estimation algorithms in resolving incident signals with different power. Further Monte Carlo experiment results prove that the classic evaluation criterion has a large misjudgment rate for resolving incident signals with different power.