基于数据驱动和统计扩散的树冠提取方法

林殷; 李恒超; 洪文

doi:10.3724/SP.J.1146.2007.00305

基于数据驱动和统计扩散的树冠提取方法

doi: 10.3724/SP.J.1146.2007.00305

林殷^{①②③ 李恒超,},
李恒超,
洪文

计量
- 文章访问数: 3369
- HTML全文浏览量: 92
- PDF下载量: 1000
- 被引次数: 0
出版历程
- 收稿日期: 2007-03-01
- 修回日期: 2007-08-08
- 刊出日期: 2008-09-19

Tree Crown Extraction Based on Data-Driven and Stochastic Diffusions

Lin Yin^{①②③ 李恒超
,},
Li Heng-Chao,
Hong Wen

摘要

摘要: 遥感图像在森林管理中有重要的作用，随着数据量的增加和分辨率的提高，从图像中提取树冠参数成为需要和可能。该文根据树冠的特征，使用标记点过程对树冠建模，采用可逆马尔科夫链蒙特卡罗算法(MCMC)配合模拟退火算法提取树冠的参数，提出新数据项以使其更好适应图像；提出数据驱动的生灭核，同时提出使用随机扩散方法代替非跳转转移核加快算法收敛速度并简化了该方法的实现。最后通过对遥感图像的实验验证了该方法的有效性。
- 图像处理 /
- 树冠提取 /
- 数据驱动的生灭 /
- 随机扩散
Abstract: Remote sensing imagery plays a key role in forestry management. The increasing availability of data and their high spatial resolution make the tree crown extraction necessary and possible. Marked point processes are employed to model the tree crowns, based on the characteristics of them. The parameters of the model are optimized by RJMCMC (Reversible Jump Markov Chain Monte Carlo sampler) and Simulate annealing. New data term is proposed to give better description of the local pattern of the tree crown; Data-driven Birth-and-Death and Stochastic Diffusions are introduced to reduce the complexity of RJMCMC kernel and accelerate convergence speed. The method is verified on remote sensing image.
- I /
- m /
- a

HTML全文

参考文献(1)

[1] Perrin G, Descombes X, and Zerubia J. Point processes inforestry: an application to tree crown detection. ResearchReport 5544, INRIA, April 2005. [2] Perrin G, Descombes X, and Zerubia J. Extraction dehouppiers par processus objet. Research Report 5037, INRIA,December 2003. [3] Ortner M, Descombes X, and Zerubia J. Improved RJMCMCpoint process sampler for object detection on images bysimulated annealing. Research Report 4900, INRIA, August2003. [4] Lafarge F, Descombes X, Zerubia J, and Deseiligny M. AParametric model for automatic 3D building reconstructionfrom high resolution Satellite Images. Research Report 5687,INRIA, September 2005. [5] Lacoste G, Descombes X, Zerubia J, and Baghdadi N.Hydrographic network extraction from radar satellite imagesusing a hierarchical model within a stochastic geometryframework Research Report 5697, INRIA, September 2005. [6] Han Feng, Tu Zhouwen, and Zhu Song-Chun. Range imagesegmentation by an Effective Jump-Diffusion method[J].IEEETrans. on Pattern Analysis and Machine Intelligence.2004,26(9):1138-1150 [7] Tu Zhouwen and Zhu Song-Chun. Image segmentation bydata-driven Markov Chain Monte Carlo[J].IEEE Trans. onPattern Analysis and Machine Intelligence.2002, 24(5):657-673 [8] Lacoste C, Descombes X, and Zerubia J. Point processes forunsupervised line network extraction in remote sensing[J].IEEE Trans. on Pattern Analysis and Machine Intelligence.2005, 27(10):1568-1579 [9] 茆诗松, 王静龙, 濮晓龙. 高等数理统计. 北京: 高等教育出版社, 1998: 419-423.Mao Shi-song, Wang Jing-long, and Pu Xiao-long. AdvancedMathematical Statistics. Beijing, China Higher EducationPress Beijing and Springer-Verlag Berlin Heidelberg, 1998:419-423.

施引文献

资源附件(0)

访问统计