ID	原文	译文
7814	仿真结果表明,利用正强化学习-正交分解算法能够更加快速地学习到最优干扰参数和最佳干扰样式,相同任务中,仅需更少的交互次数且干扰效果更好,较现有干扰策略选择算法更优。	The simulation results show that using positive reinforcement learning - orthogonal decomposition algorithm can more quickly to interfere with the optimal parameters and best interference pattern, the same task, only less interactions and interference effect is better, strategy selection algorithm is better than existing interference.
7815	针对现有多目标火力分配(weapon target assignment,WTA)方法很难适用于不确定情况下防空反导作战的问题,提出了基于模糊多目标规划的防空反导WTA方法。	In view of the existing target fire distribution (weapon target the assignment, WTA) method is difficult to apply to the uncertain circumstances anti-air warfare problem, put forward the anti-air WTA method based on fuzzy multi-objective programming.
7816	首先,采用三角模糊数刻画不确定的目标威胁度,在考虑防空反导作战特点的基础上,基于模糊多目标规划建立了WTA模型;	First of all, the triangular fuzzy number to depict uncertainty of target threat degree, considering the anti-air combat characteristics, on the basis of WTA model was built based on fuzzy multi-objective programming;
7817	然后,根据必要性测度原理将含有模糊参数的目标函数进行了等价清晰化;	Then, according to the principle of necessity measure will contain the objective function of fuzzy parameters equivalent motivation;
7818	接着,提出了具有单/双势阱的多目标量子行为粒子群算法用于求解WTA模型,该算法采用了单/双势阱位置更新方式、粒子混合随机变异方法、领导粒子两阶段选取方法;	Then, put forward with single/double trap multi-objective quantum behavior of the particle swarm optimization (pso) algorithm is used to solve WTA model, the algorithm adopts the single/double potential well location updating method, particle mixing random variation method, leading particle two phase selection method;
7819	最后,通过实例仿真验证了模型的合理性和算法的有效性。	Finally, an example simulation validate the rationality and validity of the model.
7820	为了充分利用高光谱图像邻域像元间的相似性与独特性这一特征信息,提出了一种基于核函数的联合稀疏表示分类方法(kernel joint sparse representation classification,K-JSRC)来提高高光谱图像的分类精度。	In order to make full use of hyperspectral images, the similarity between neighborhood pixels, and the uniqueness of this characteristic information, this paper proposes a joint sparse representation classification method based on kernel function (kernel to be sparse representation classification, K - JSRC) to improve the classification accuracy of hyperspectral image.
7821	该方法通过一种改进的核函数对每个待测中心像元的所有邻域像元自适应的予以不同权重来测量待测中心像元与邻域像元的相似度从而得到不规则的最优邻域窗口。	This method by an improved kernel function to each under test center like yuan all the neighborhood pixels, adaptive to different weights to measure under test center as yuan and the neighborhood pixels, the optimal neighborhood similarity and irregular window.
7822	在Indian Pines和University of Pavia两个高光谱数据集上的实验结果表明,提出的分类算法对高光谱图像进行了很好的分类并且其分类精度优于同类算法。	In Indian Pines and the University of Pavia two hyperspectral data sets on the experimental results show that the proposed classification algorithm had good classification of hyperspectral image and its classification precision is superior to the similar algorithm.
7823	针对现有随机有限集(random finite set,RFS)扩展目标滤波器不能输出航迹的问题,提出了基于标签RFS滤波器的多扩展目标跟踪算法。	In view of the existing stochastic finite set (random finite set, RFS) extended target filter cannot output track problem, put forward the more extended target tracking algorithm based on label RFS filter.