HE Yu,YANG Fu-yi,YANG Lu.A Method of Small Target Detection in the Starry Sky Background based on NSCT Distance Map[J].Journal of Chengdu University of Information Technology,2017,(03):264-268.[doi:10.16836/j.cnki.jcuit.2017.03.006]
基于NSCT距离像的星空背景小目标检测技术
- Title:
- A Method of Small Target Detection in the Starry Sky Background based on NSCT Distance Map
- 文章编号:
- 2096-1618(2017)03-0264-05
- 关键词:
- 星空背景; 非下采样Contourlet变换; 方向子带; 距离像
- Keywords:
- in the starry sky background; NSCT; direction band; distance map
- 分类号:
- TP391.41
- 文献标志码:
- A
- 摘要:
- 点状空间目标和长条状背景星存在几何特征的不同,在频域中表现为:长条状背景星主要分布某方向子带上,而点状目标在所有方向子带上的分布近似相同。非下采样Contourlet(NSCT)变换可以对图像进行灵活的多尺度、多方向和平移不变性的分解,能很好地捕捉长条状轮廓,文中提出的基于NSCT距离像的目标检测技术中将某尺度上各方向分量间的比值作为特征向量,通过计算它与中心向量的距离得到距离像,从距离像的统计直方图中可以自适应地选择分割点状目标的门限,实验证明该方法能有效地检测图像中的点状小目标。
- Abstract:
- There are some differences of geometry features between the points-like small target and the strip-like background stars. In the view of frequency analysis, the background stars will have a main direction band in the frequency domain, because it has aStrip-shaped look, while the points-like small target can be treated as pulse signal, so its energy will distribute all the band of frequency domain. The Nonsubsampling Contourlet Transform(NSCT)hasflexible multiscale, multidirection, and shift-invariant image decomposition characteristics, so it can efficiently capture the differences of the geometry features. This paper proposes a detection method based on the NSCT Distance Map: defined the ratios between each direction component as a feature vector, and then get a distance image by calculating the distance between the feature vector and the center vector, after that, can adaptively get the threshold that used to segmented the small target from the image from the grayscale histogram. Experiments show that the method can efficiently detect the small target from the image.
参考文献/References:
[1] 魏敦生.星空背景下小目标的跟踪与识别[D].上海:上海交通大学,2012.
[2] 魏芳.基于多尺度几何分析的红外弱小目标检测方法研究[D].成都:电子科技大学,2012.
[3] 孟威.复杂背景自适应红外小目标检测[D].杭州:浙江大学,2013.
[4] 关雪梅.基于空域的图像增强技术研究[J].赤峰学院学报(自然科学版),2012,(8):22-24.
[5] 修鹏山.深空背景下小目标检测和跟踪[D].北京:中国科学院大学,2015.
[6] 郭晓军,万龙,刘峰.基于扩展卡尔曼滤波的空间小目标跟踪算法[J].电光与控制,2016,(4):57-61.
[7] 黄宗福,孙刚,陈曾平.空间目标检测序列图像中恒星目标抑制[J].中国图像图形学报,2013,(7):799-804.
[8] Ha-ning X,S-y Z,Ping-yuan C.Deep space moving target detection based on technology of image matching[J].Control and Decision Conference:IEEE,2014:3294-9.
[9] 杨馥溢,杨璐,魏敏,等.星空背景的多运动小目标检测方法[J].成都信息工程大学学报,2016,31(6).
[10] 吴一全,罗子娟,吴文怡.基于NSCT的红外图像小目标检测技术[J].中国图像图形学报,2009,(3):477-81.
[11] Do MN,Vetterli M.Framing pyramids.Signal Processing[J].IEEE Transactions on,2003,51(9):2329-42.
[12] Do M N,Vetterli M.Contourlets:A directional multiresolution image representation[J].2001,(1):357-360.
[13] Da Cunha AL,Zhou J,Do MN.The nonsubsampled contourlet transform:theory,design,and applications.Image Processing[J].IEEE Transactions on.2006,5(10):3089-101.
[14] Do MN,Vetterli M.The contourlet transform:an efficient directional multiresolution image representation.Image Processing[J].IEEE Transactions on.2005,14(12):2091-106.
[15] 盛文,邓斌,柳健.一种基于多尺度距离像的红外小目标检测方法[J].电子学报,2002.
备注/Memo
收稿日期:2017-06-27 基金项目:四川省教育厅基金资助项目(14ZA0176)