ID	原文	译文
5864	为了改善遥感图像超分辨重建(super-resolution reconstruction,SRR)效果,针对以往仅适用于单特征空间的稀疏字典超分辨算法,提出同时适用于两个特征空间的双参数Beta过程联合字典(Beta process joint dictionary,BPJD)遥感图像SRR方法。	In order to improve the remote sensing image super resolution reconstruction (super - resolution reconstruction, SRR) effect, the past is only applicable to single feature space of sparse dictionaries super resolution algorithm, is put forward at the same time is suitable for double parameters of two feature space Beta process joint dictionary (Beta process to be a dictionary, BPJD) remote sensing image SRR method.
5865	首先,根据遥感图像退化模型生成训练样本图像,并分别对高、低分辨率图像进行分块和Gibbs采样,生成字典训练样本。	First of all, according to remote sensing image degradation model to generate training samples, and high and low resolution image partition and Gibbs sampling, to generate a dictionary training sample.
5866	然后,依据BPJD,建立连接高、低分辨率遥感图像空间的双参数联合稀疏字典,将字典稀疏系数分解为系数权值和字典原子的乘积。	Then, on the basis of BPJD, high connection is established, the low spatial resolution remote sensing image of double parameters in combination with sparse dictionary, the dictionary sparse coefficient is decomposed into the product of the coefficient of weight and a dictionary of atoms.
5867	依据字典原子指标训练和更新字典,得到高低分辨率联合字典映射矩阵。	According to the dictionary atoms index training and update the dictionary, have high and low resolution joint dictionary mapping matrix.
5868	最后,进行遥感图像超分辨稀疏重构。	Finally, through super resolution remote sensing image sparse reconstruction.
5869	实验结果表明:所提方法可自适应地缩小字典尺寸,能以更小尺寸的稀疏字典重建更高质量的超分辨遥感图像,重建结果图像的纹理细节信息更丰富,峰值信噪比和结构相似性度均有提高。	Experimental results show that the proposed method can decrease adaptively the dictionary size, to a smaller size of the sparse dictionary to reconstruct the super resolution remote sensing images of higher quality of reconstructed image texture detail information richer, peak signal-to-noise ratio and the structure similarity degrees were increased.
5870	针对存在主用户模拟攻击(primary user emulation attack,PUEA)下的频谱感知问题,提出了一种基于攻击强度阈值(attack-aware threshold,ART)的协作频谱感知方法。	Main users against simulated attacks (the primary user emulation attack, PUEA) under the spectrum sensing problem, this paper proposes a threshold based on attack power (attack - aware threshold, ART) collaborative spectrum sensing method.
5871	该方法首先建立了存在PUEA的网络模型,然后分析4种状态下协作频谱感知的特点,采用检测统计量的一阶和二阶矩估计出PUEA攻击的概率。	The method exists PUEA network model is established first, and then analysis the characteristics of the collaborative spectrum sensing, four kinds of condition with first and second order moments of the test statistics estimate PUEA attack probability.
5872	最后以总误差概率为目标函数,求出不同攻击强度下的最佳门限阈值,并通过ART方法完成在不同攻击强度下的协作频谱感知。	Finally to total error probability as the objective function, the optimum threshold of under different attack power threshold, and through the ART method in collaborative spectrum sensing under different attack power.
5873	仿真实验表明,该方法能够估计恶意信号出现的概率,有效减少频谱感知总误差概率,提高整个网络的频谱感知性能。	Simulation experiments show that this method can estimate the probability of malicious signal appears, effectively reduce the total error probability spectrum perception, improve the spectrum sensing performance of the entire network.