摘要
该文首先提出了非相干平均、多分辨率分解、富氏变换的特征提取方法;然后用隐Markov模型来描述、识别基于一维像的特征序列。3类ISAR飞机目标的实测数据用来验证上述方法的有效性,取得了99.
A method of feature extraction of incoherent average, multiresolution decomposition, Fourier transformation is proposed. The hidden Markov model (HMM) technique is used to identify feature sequences resulting from range profile sequences. A data set of real radar range profile of three class aircrafts gotten from an ISAR is used to test the effectiveness of above methods. Identification accuracy of 99.33% is obtained.
出处
《南京理工大学学报》
EI
CAS
CSCD
1998年第3期224-227,251,共5页
Journal of Nanjing University of Science and Technology
关键词
雷达
目标识别
特征提取
隐MARKOV模型
radar, target identification, range profile, feature extraction
hidden Markov model