ID 原文 译文
54147 最后通过极大值参数估计方法实现对旋翼转速、叶片长度的估计。 and finally the rotor speed and blade length can be estimated by the maximum value parameter estimation method.
54148 结果表明RSP-CFD方法对旋翼无人机微动特征的提取具有较高的准确性，弥补了传统方法的不足，进而为旋翼无人机的分类提供理论基础和技术支撑。 The results show that RSP-CFD method has high accuracy in extracting the micro-motion features of the rotor UAV, which makes up for the shortcomings of the traditional methods, and then provides theoretical foundation and technical support for the classification of the rotor UAV.
54149 针对传统感知算法在低信噪比时检测性能低和深度学习感知算法网络训练量大、复杂度高等问题，本文提出一种在均值辅助下的长短时记忆网络(Long Short-Term Memory，LSTM)频谱感知算法。 Due to the low detection performance of traditional sensing algorithms at low signal-to-noise ratio regions and the large amount of network training and high complexity of deep learning sensing algorithms, this paper proposes a mean-assisted LSTM network spectrum sensing algorithm.
54150 首先对接收信号序列做多点均值计算， This paper first calculates the multi-point average of the received signal,
54151 然后利用所得的均值构造特征向量并作为LSTM网络的输入来训练网络， then constructs the feature vector using the calculated average as the LSTM network input to train the network,
54152 最后利用训练好的网络对新的接收序列进行感知。 and finally senses the available spectrum using the trained network.
54153 仿真结果表明：相比于传统算法，所提算法在检测性能上有较大提升; 相对于利用原始接收序列直接训练的深度学习算法，所提算法的复杂度大幅下降。 The simulation results show that the detection performance of the proposed algorithm outperforms that of the traditional algorithms, and that the proposed algorithm can achieve lower complexity than the deep learning algorithm trained with the original received sequence.
54154 结合空间信息约束的高光谱稀疏解混技术是高光谱图像稀疏解混领域的研究热点之一。 Sparse Unmixing(SU) combined with spatial information constraints is one of the research hotspots in the field of Hyperspectral Unmixing(HU).
54155 为了克服高光谱图像在自然场景中的空间结构难以精确表示的缺点，本文提出了一种多尺度光谱相似性指导的高光谱解混算法。 In order to overcome the shortcomings that the spatial structure of hyperspectral images in natural scenes is difficult to accurately represent. Hence, a Hyperspectral Unmixing algorithm guided by multiscale spectral similarity was proposed in this paper.
54156 首先，将高光谱图像分割成具有空间结构的近似域光谱图像; First, segment the Hyperspectral Image into an approximate domain spectral image with a spatial structure.