我们的网站为什么显示成这样?

可能因为您的浏览器不支持样式,您可以更新您的浏览器到最新版本,以获取对此功能的支持,访问下面的网站,获取关于浏览器的信息:

|Table of Contents|

基于SDA和CART算法的面向对象分类研究(PDF/HTML)

《南京林业大学学报(自然科学版)》[ISSN:1000-2006/CN:32-1161/S]

Issue:
2015年03期
Page:
6-12
Column:
专题报道
publishdate:
2015-05-30

Article Info:/Info

Title:
Object-oriented classification based on SDA and CART algorithm
Article ID:
1000-2006(2015)03-0006-07
Author(s):
ZHANG Yangyang12 LIU Haijuan12 ZHANG Ting12 XU Yannan12* SHI Hao3
1.College of Forestry, Nanjing Forestry University,Nanjing 210037,China;
2.Co-Innovation Center for the Sustainable Forestry in Southern China, Nanjing Forestry University, Nanjing 210037, China;
3.Jiangsu Provincial Environmental Monitoring Center, Nanjing 210036, China
Keywords:
object-oriented classification SDA CART model sensitive analysis
Classification number :
TP75
DOI:
10.3969/j.issn.1000-2006.2015.03.002
Document Code:
A
Abstract:
Use the QuickBird remotely sensed imagery in 2005 of Futian District, Shenzhen City as the main data source, the spectrum, shape and texture features of different land-use were extracted by the object-oriented multiresolution segmentation. Then we used stepwise discriminant analysis(SDA)with classification and regression trees(CART)to build “multi-segmentation, multi-features” classification model. The results showed that:(1)With SDA, it could choose an advanced feature set. In this study, among the 32 features we totally chose 27 important features and sorted them by the importance in tables. Among them, the spectrum and texture features were mostly sorted forward, while the shape features were sorted later except the LengthWidth was ranked fifth.(2)The CART model could recognize more important features, and calculated separation thresholds for object-oriented classification automatically. In this study, Mean_b3, LengthWidth, Ratio_b3, GLDVE and NDVI features were the important classified nodes.(3)Under the condition of improving or no significantly reducing classification accuracy, SDA combined with CART model achieved the feature dimension reduction. When we chose R2>0.2, the CART16 model obtained an optimal result of training precision 94.44% and test precision 83.37%, while the feature set was the smallest, compared with primary one, the number of features decreased to half.

References

[1] 郑雅敏. 基于遗传算法的特征选择方法的改进研究[D]. 重庆:重庆大学, 2008.Zheng Y M. Improvement of feature selection method based on genetic algorithm[D]. Chongqing:Chongqing University,2008.
[2] Kononenko I. Estimation attributes: analysis and extensions of RELIEF [C]// Machine Learning: The 1994 European Conference on Machine Learning. Catania, 1994:174-182.
[3] John G H, Kohavi R, Pfleger K. Irrelevant features and the subset selection problem[C]// Machine Learning: Proc of the 11th Int Conf on Machine Learning. New Brunswick, 1994: 121-129.
[4] 姚旭, 王晓丹, 张玉玺, 等. 特征选择方法综述[J]. 控制与决策, 2012, 27(2): 161-166.Yao X, Wang X D, Zhang Y X, et al.Summary of feature selection algorithms[J]. Control and Decision,2012,27(2):161-166.
[5] 吴浩苗, 尹中航, 孙富春. Relief 算法在笔迹识别的应用[J]. 计算机应用, 2006, 26(1): 174-177.Wu H M, Yi Z H, Sun F C.Application of relief in handwriting recognition[J]. Journal of Computer Applications, 2006, 26(1): 174-177.
[6] 张克军, 刘哲. 基于改进的 FCM 和 KPCA 的多光谱图像特征提取方法[J]. 科学技术与工程, 2007, 7(8): 1657-1661.Zhang K J, Liu Z. Feature extraction for multispectral images based on improved FCM and KPCA [J]. Science Technology and Engineering, 2007,7(8): 1657-1661.
[7] 余晓敏, 湛飞并, 廖明生, 等. 利用改进 SEaTH 算法的面向对象分类特征选择方法[J]. 武汉大学学报:信息科学版, 2012, 37(8): 921-924. Yu X M, Zhan F B, Liao M S, et al. Object-oriented feature selection algorithms based on improved SEaTH algorithms [J]. Geomatics and Information Science of Wuhan University, 2012, 37(8): 921-924.
[8] Ruiliang Pu, Shawn Landry. A comparative analysis of high spatial resolution IKONOS and WorldView-2 imagery for mapping urban tree species[J]. Remote Sensing of Environment, 2012,124: 516-533.
[9] 潘琛, 林怡, 陈映鹰. 基于多特征的遥感影像决策树分类[J]. 光电子·激光, 2010, 21(5): 731-736. Pan C, Lin Y, Chen Y Y.Decision tree classification of remote sensing images based on multi-feature[J]. Journal of Optoelectronics Laser, 2010,21(5):731-736.
[10] 唐银凤, 黄志明, 黄荣娟, 等. 基于多特征提取和 SVM 分类器的纹理图像分类[J]. 计算机应用与软件, 2011, 28(6): 22-25.Tang Y F, Huang Z M, Huang R J, et al. Texture image classification based on multi-feature extraction and svm classifier[J]. Computer Applications and Software, 2011, 28(6): 22-25.
[11] 吴芳, 贾永红. 基于 PCA—BPNN 的多光谱遥感影像分类[J]. 地理空间信息, 2006, 4(1): 15-17.Wu F, Jia Y H. Classification of multispectral remote sensing image based on PCA-BPNN [J]. Geospatial Information, 2006, 4(1): 15-17.
[12] 吴红霞, 吴悦, 刘宗田, 等. 基于 Relief 和 SVM-RFE 的组合式 SNP 特征选择[J]. 计算机应用研究, 2012, 29(6):2074-2077. Wu H X, Wu Y, Liu Z T, et al. Combined SNP feature selection based on relief and SVM-RFE[J]. Application Research of Computer, 2012, 29(6): 2074-2077.
[13] 张颖娜,张超,赵冬玲,等. 利用ReliefFO和海明遗传算法的面向对象分类特征选择算法[J]. 武汉大学学报:信息科学版,2010,35(12): 1444-1448.Zhang Y N, Zhang C, Zhao D L, et al. Object-oriented feature selection algorithms with ReliefFO and Haiming genetic algorithm[J]. Geomatics and Information Science of Wuhan University, 2010, 35(12): 1444-1448.
[14] 潘洁, 李明诗. 基于信息量的高分辨率影像纹理提取的研究[J]. 南京林业大学学报:自然科学版, 2010, 34(4):129-134.Pan J, Li M S. Textural features analysis of high-resolution remote sensing image based on the information abundance[J]. Journal of Nanjing Forestry University:Natural Sciences Edition,2010,34(4):129-134.
[15] 王登峰, 杨志刚, 魏安世. 纹理信息在遥感影像分类中的应用[J]. 南京林业大学学报:自然科学版, 2010, 34(3):97-100.Wang D F, Yang Z G, Wei A S. Application of texture information on classification of remote sensing imagery[J]. Journal of Nanjing Forestry University: Natural Sciences Edition,2010,34(3): 97-100.
[16] Martha T R, Kerle N, van Westen C J, et al. Segment optimization and data-driven thresholding for knowledge-based landslide detection by object-based image analysis[J]. IEEE Transactions on Geoscience and Remote Sensing, 2011, 49(12):4928-4943.
[17] Stumpf A, Kerle N. Object-oriented mapping of landslides using Random Forests[J]. Remote Sensing of Environment, 2011, 115(10): 2564-2577.
[18] Mladinich C S. An evaluation of object-oriented image analysis techniques to identify motorized vehicle effects in semi-arid to arid ecosystems of the American west[J]. GIS Science & Remote Sensing, 2010, 47(1): 53-77.

Last Update: 2015-05-30