ID 原文 译文
14185 首先,利用皮尔逊相关系数计算证据体相关性,并由此定义证据权重计算平均证据。 Firstly‚Pearson correlation coefficient is used to calculate the correlation between evidence bodies‚and the weight of evidence is defined to calculate the average evidence.
14186 然后,利用平均证据计算各证据焦元距离,根据焦元距离将冲突重新分配给各焦元的BPA,构造新证据体。 Then‚the average evidence is used to calculate the focal distance of each evidence element‚and the conflict is redistributed to each focal element BPA according to the focal distance to construct a new evidence body.
14187 最后,对新证据体进行传统Dempster组合,得到融合结果。 Finally‚the traditional Dempster combination of the new evidence body is conducted to obtain the fusion result.
14188 仿真实验表明,新方法在处理高冲突证据时准确度高、收敛速度快,较好地解决了高冲突证据融合问题。 Simulation experiments show that the new
14189 针对远距离压制干扰下机动目标跟踪问题,分析了远距离大功率压制干扰对于雷达探测及测量精度的影响,引入新的量测模型模拟雷达在压制干扰下由于探测概率下降出现的目标暂消现象,并利用去相关无偏量测转换方法将球坐标下的量测转换到直角坐标系下。 To realize the tracking of maneuvering targets in the presence of long-range suppression jamming‚the influence of long-range‚high-power suppression jamming on radar detection and measurement accuracy is analyzed. A new measurement model is introduced to simulate the temporary disappearance of the targets caused by the decrease of detection probability of the radar in the presence of suppression jamming. The measurement in spherical coordinates is converted to measurement in right angle coordinates by using the decorrelated and unbiased measurement conversion method.
14190 在此基础上,基于自适应联邦滤波的思想建立了雷达压制干扰下的机动目标跟踪算法。 Then‚a maneuvering target tracking algorithm under suppression jamming on the radar is established based on the adaptive federal filter.
14191 仿真结果表明,在远距离压制干扰下,所提方法可以保持对目标在多种机动情况下的稳定、准确跟踪,与现有方法相比具有明显的优势。 The simulation results show that‚the proposed method can maintain stable and accurate tracking of targets in a variety of maneuvering situations under long-range suppression jamming‚which is obviously superior to the existing methods. This method has high accuracy and fast convergence rate when dealing with high-conflict evidence‚which provides a good solution to the problem of high-conflict evidence fusion.
14192 军事目标分类是一个重要的研究方向。 It is an important research direction to classify military images.
14193 在复杂背景下不同的军事目标的相似度较高,使得基于传统视觉特征的军事目标的分类精度不高。 The classification accuracy based on traditional visual features is not high due to the high similarity of different military targets in complex background.
14194 提出一种基于改进典型相关分析的局部二值模式(LBP)和分层梯度方向直方图相结合的军事目标分类方法。 A method of military image classification is proposed by combining Local Binary Pattern(LBP) with Pyramid Histogram of Oriented Gradients(PHOG) through Local Discriminant Canonical Correlation Analysis(LDCCA).