第38卷第5期画消防车>一个人的精彩
2019年10月
Vol. 38, No. 5October, 2019红外与毫米波学报J. Infrared Millim. Waves 文章编号：1001 -9014(2019)05-0578-09DOI ： 10.11972/j. issn. 1001 -9014.2019.05.006
Rotation-invariant infrared aerial target identification bad on SRC
JIN Lu 1'2'3, LI Fan-Ming 13*
三重境界>关公战长沙, LIU Shi-Jian 13, WANG Xiao 1'2'3Receiveddate ： viddate ： 2019-07-08 收稿日期：2019-01-09，修回日期：2019-07-08Foundation items ： Supported by the Thirteen Five National Defen Rearch Foundation (Jzx2016-0404/Y72-2) ; Shanghai Key Laboratory of Criminal Scene Evidence funded Foundation (2017xcwzk08).
Biography :JIN Lu (1991-), female , Wuhan, PhD. Rearch area involves computer vision and pattern recognition. E-mail:jinlu0716@163. com.* Corresponding author ：E-mail ： lfmjws@163. com (1. Shanghai Institute of Technical Physics , Chine Academy of Sciences , Shanghai 200083, China;
2. University of Chine Academy of Sciences , Beijing 100049, China ;
3. CAS Key Laboratory of Infrared System Detection and Imaging Technology , Shanghai Institute of Technical Physics ,
Shanghai 200083, China)
Abstract ： Aircraft identification is implemented on thermal images acquired from ground-to-air infrared cameras. SRC is proved to be an effective image classifier robust to noi, which is quite suitable for thermal image tasks. However, rotation invariance is challenging requirements in this task. To solve this issue, a method is propod to compute the target main orientation firstly, then rotate the target to a reference direction. Secondly, an over complete dictionary is learned from histogram of oriented gradient features of the rotated targets. Thirdly , a spar reprentation model is introduced and the identification problem is converted to a -minimization prob lem. Finally , different aircraft types are predicted bad on an evaluation index , which is called residual error. To validate the aircraft identification method , a recorded infrared aircraft datat is implemented in an airfield. Experimental results show that the propod method achieves 98. 3% accuracy , and recovers the identity beyond 80% accuracy even when the test images are corrupted at 50%.
Key words : infrared image , aircraft identification, rotation invariant, spar reprentation classification PACS : 42. 30. Sy , 07. 05. Mh
互动哥基于稀疏表示的红外空中目标分类算法
金駅23，李范計3；刘士建1,3，王霄1,2,3
王维的介绍(1.中国科学院上海技术物理研究所，上海200083;
2.中国科学院大学，北京100049;
3.中国科学院红外探测与成像技术重点实验室，上海200083)
摘要：针对红外空中目标，提出了一种基于稀疏表示的快速分类算法.该工作的技术难点表现在训练样本较 少，算法需要具有旋转不变性、较高的抗噪性和实时性.针对这些难点，首先根据红外空中面目标的梯度信息 和统计特性，计算出图像主方向，然后将主方向旋转至同一参考方向.接着基于稀疏表示原理，把分类问题转 化为1范数最小化问题，最后用快速收敛方法得到分类结果.实验结果表明该方法能够达到98.3%的正确率， 给测试图像50%的像素叠加噪声后，分类正确率仍大于80%.
关键词：红外图像；空中目标；旋转不变性；稀疏表示分类
中图分类号:TP391.4 文献标识码:A
Introduction
南瓜怎么蒸好吃Infrared target recognition and classification are sig
nificant parts in video surveillance and aeronautics appli
cations. In aeronautics applications , aircraft are the main targets to surveil. Especially in ground-to-air appli cations ,a system which has good performance at anti jamming, fast identification friend or foe and stable track
上海车牌拍卖流程
