针对空间目标识别中特征提取难、准确率低等问题,提出了一种基于雷达高分辨率距离像(high range resolution profile,HRRP)时频特征和多尺度非对称卷积神经网络的目标识别算法。采用离差标准化、多特显点绝对对齐消除目标的强度敏感性...针对空间目标识别中特征提取难、准确率低等问题,提出了一种基于雷达高分辨率距离像(high range resolution profile,HRRP)时频特征和多尺度非对称卷积神经网络的目标识别算法。采用离差标准化、多特显点绝对对齐消除目标的强度敏感性和平移敏感性,利用雷达多普勒测速数据消除目标高速运动对HRRP产生的展宽、畸变、波峰分裂等影响。对HRRP进行时频分析,提取其时频特征。通过不同尺度的非对称卷积,实现时频特征不同精细程度和不同方向的特征提取。实测数据处理结果表明,文中方法目标识别准确率高,而且在同平台目标识别、抗姿态敏感性等方面具有很好的效果。展开更多
针对多极化高分辨率一维距离像(high range resolution profile,HRRP)在目标识别过程中存在计算量和数据量大、识别算法复杂的问题,提出一种基于Bagging-SVM动态集成的目标识别方法。该方法首先提取多极化HRRP平移不变特征向量,然后运用...针对多极化高分辨率一维距离像(high range resolution profile,HRRP)在目标识别过程中存在计算量和数据量大、识别算法复杂的问题,提出一种基于Bagging-SVM动态集成的目标识别方法。该方法首先提取多极化HRRP平移不变特征向量,然后运用Bagging方法结合基于动态互信息的特征选择方法生成基分类器,最后引入基分类器差异度进行选择性集成。实验验证该方法在缩减数据规模和计算量的同时,能有效利用多极化特征信息,得到较高的分类正确率,并且松弛了HRRP目标的姿态敏感性。展开更多
The principle and method of both radar target imaging and velocity measurement simultaneously based on step frequency waveforms is presented. Velocity compensation is necessary in order to obtain the good High resolut...The principle and method of both radar target imaging and velocity measurement simultaneously based on step frequency waveforms is presented. Velocity compensation is necessary in order to obtain the good High resolution range profile since this waveform is greatly sensitive to the Doppler shift. The velocity measurement performance of the four styles is analyzed with two pulse trains consisted of positive and negative step frequency waveforms. The velocity of targets can be estimated first coarsely by using the pulse trains with positive-positive step frequency combination, and then fine by positive-negative combination. Simulation results indicate that the method can accomplish the accurate estimation of the velocity with efficient computation and good anti-noise performance and obtain the good HRRP simultaneously.展开更多
For radar high resolution range profile (HRRP) recognition, three aspects are of great importance to improve the performance, i.e. discrimination for outlier, classification for inner and an accurate description for f...For radar high resolution range profile (HRRP) recognition, three aspects are of great importance to improve the performance, i.e. discrimination for outlier, classification for inner and an accurate description for feature space. To tackle these issues, a novel target recognition method is designed, denoted by the multiple support vectors (multi-SV) method. With the proposed method, a special framework is constructed by a treble correlate support vector model to segment the feature space to two regions with the distribution of density, and then the description and classification hyperplane for each region are achieved. Based on the support vector framework, this method needs less memory and computation complexity to fit practical radar HRRP recognition. Finally, the experiment based on the measured data verifies the excellent performance of this method.展开更多
This paper studies a detection method of targets of high resolution radar operating at the band of millimeter-wave(32-38GHz) under the background of the clutters, and proposes a new nonparametric detection method, whi...This paper studies a detection method of targets of high resolution radar operating at the band of millimeter-wave(32-38GHz) under the background of the clutters, and proposes a new nonparametric detection method, which not only does less computation, but also is able to detect multiple extended targets radially distributed along distance "corridor", based on the position (range) correlation information of one-dimensional range images(or called range profiles) of high resolution radar targets. The experimental results, on the real echo data of tank illuminated by the millimeter-wave stepped frequency high resolution radar, have certified that such a method presented in this paper is a very effective detection method for multiple extended targets.展开更多
文摘针对空间目标识别中特征提取难、准确率低等问题,提出了一种基于雷达高分辨率距离像(high range resolution profile,HRRP)时频特征和多尺度非对称卷积神经网络的目标识别算法。采用离差标准化、多特显点绝对对齐消除目标的强度敏感性和平移敏感性,利用雷达多普勒测速数据消除目标高速运动对HRRP产生的展宽、畸变、波峰分裂等影响。对HRRP进行时频分析,提取其时频特征。通过不同尺度的非对称卷积,实现时频特征不同精细程度和不同方向的特征提取。实测数据处理结果表明,文中方法目标识别准确率高,而且在同平台目标识别、抗姿态敏感性等方面具有很好的效果。
文摘针对多极化高分辨率一维距离像(high range resolution profile,HRRP)在目标识别过程中存在计算量和数据量大、识别算法复杂的问题,提出一种基于Bagging-SVM动态集成的目标识别方法。该方法首先提取多极化HRRP平移不变特征向量,然后运用Bagging方法结合基于动态互信息的特征选择方法生成基分类器,最后引入基分类器差异度进行选择性集成。实验验证该方法在缩减数据规模和计算量的同时,能有效利用多极化特征信息,得到较高的分类正确率,并且松弛了HRRP目标的姿态敏感性。
文摘The principle and method of both radar target imaging and velocity measurement simultaneously based on step frequency waveforms is presented. Velocity compensation is necessary in order to obtain the good High resolution range profile since this waveform is greatly sensitive to the Doppler shift. The velocity measurement performance of the four styles is analyzed with two pulse trains consisted of positive and negative step frequency waveforms. The velocity of targets can be estimated first coarsely by using the pulse trains with positive-positive step frequency combination, and then fine by positive-negative combination. Simulation results indicate that the method can accomplish the accurate estimation of the velocity with efficient computation and good anti-noise performance and obtain the good HRRP simultaneously.
文摘For radar high resolution range profile (HRRP) recognition, three aspects are of great importance to improve the performance, i.e. discrimination for outlier, classification for inner and an accurate description for feature space. To tackle these issues, a novel target recognition method is designed, denoted by the multiple support vectors (multi-SV) method. With the proposed method, a special framework is constructed by a treble correlate support vector model to segment the feature space to two regions with the distribution of density, and then the description and classification hyperplane for each region are achieved. Based on the support vector framework, this method needs less memory and computation complexity to fit practical radar HRRP recognition. Finally, the experiment based on the measured data verifies the excellent performance of this method.
文摘This paper studies a detection method of targets of high resolution radar operating at the band of millimeter-wave(32-38GHz) under the background of the clutters, and proposes a new nonparametric detection method, which not only does less computation, but also is able to detect multiple extended targets radially distributed along distance "corridor", based on the position (range) correlation information of one-dimensional range images(or called range profiles) of high resolution radar targets. The experimental results, on the real echo data of tank illuminated by the millimeter-wave stepped frequency high resolution radar, have certified that such a method presented in this paper is a very effective detection method for multiple extended targets.
