基于决策树技术的遥感影像分类研究
|
摘要
决策树分类法已被应用于许多分类问题，但应用于遥感分类的研究成果并不多见。决策树分类法具有灵活、直观、清晰、强健、运算效率高等特点，在遥感分类问题上表现出巨大优势。本文以广东省广州市从化地区的SPOT5卫星遥感影像为研究对象，基于决策树分类算法在遥感影像分类方面的深厚潜力，探讨了6种不同的决策树算法——包括单一决策树模型(CART，CHAID，exhaustive CHAID，QUEST)和组合决策树模型(提升树，决策树森林)。首先对决策树算法结构、算法理论进行了阐述，然后利用这些决策树算法进行遥感土地覆盖分类实验，并把获得的结果与传统的最大似然分类和人工神经元网络分类进行比较。结果表明：
1．在卫星影像的整体分类精度的上，决策树分类技术要优于神经元网络和最大似然分类。相对于神经元网络方法，决策树在训练样本数据的速度要快，并且执行效率要高，对于输入数据空间特征和分类标识，具有更好的弹性和鲁棒性。并且，决策树并不象神经网络方法那样具有黑箱(black box)结构，和分析人员有着良好的交互性和透明性，而神经元网络初始参数的设定是根据实践经验来确定的，缺乏有效的理论指导，带有很强的主观性，并且在训练过程中这些参数要经过不断的调整，才能生成一个较好的网络模型，而且，神经网络的训练非常费时并且对训练结果的好坏事先缺乏判断，这些都是不如决策树分类的地方(在本研究中，虽然神经元网络的分类精度达到了82％，要优于CHAID，但是这种分类精度是在不断调整神经网络的初始参数通过数次尝试才达到的)。相对于最大似然分类，决策树的树状分类结构对数据特征空间分布不需要预先假设某种参数化密度分布，所以其总体分类精度优于传统的参数化统计分类方法。
2．在所有的决策树分类技术中，，组合决策树模型(决策树森林和TreeBoost)的总体表现要比单一决策树模型(CART，CHAID，Exhaustive CHMD，QUEST)模型优秀，但是组合决策树模型的“黑箱”结构其可视性和可解释性又不如单一决策树，正以为如此，在分类过程中要选择那种模型要视情况而定，单一树模型能直观的理解预测变量在分类中的作用和能生成清晰的类别判别规则，而组合树模型通常能比单一树模型得到更高的预测精度。
3．在组合决策树模型中，TreeBoost是通过将每次预测函数的输出赋以一定权重并重复的应用该预测函数来使得总预测误差达到最小化而提高分类精度的。而与TreeBoost模型不同的是，决策树森林中的每棵树都是独立平衡生长的并且它们在所有的树生成之前是不相互影响的，整个森林的预测精度是由其中每一棵树的预测精度组合而得到的。相对于单一树模型，组合树模型能显著提高分类精度，并且该模型能避免过拟合现象，因而不需要对其进行修剪。一般来说，组合树模型树的数目越多，该模型的预测效果越好。
4．在单一决策树模型中，各分类方法的差别主要体现在决策树生长过程中预测变量的选择和变量分割点的选择。在本研究中，CART(精度86.5％)要优于QUEST(82.7％)＞CHAID(81.5％)＞Exhausitve CHAID(81.3％)。
Choice of a classification algorithm is generally based upon a number of factors, among which are availability of software, ease of use, and performance, measured here by overall classification accuracy. The maximum likelihood (ML) procedure is, for many users, the algorithm of choice because of its ready availability and the fact that it does not require an extended training process. Artificial neural networks (ANNs) are now widely used by researchers, but their operational applications are hindered by the need for the user to specify the configuration of the network architecture and to provide values for a number of parameters, both of which affect performance. The ANN also requires an extended training phase.
In the past few years, the use of decision tree (DTs) to classify remotely sensed data has increased. Proponents of the method claim that it has a number of advantages over the ML and ANN algorithms. The DT is computationally fast, make no statistical assumptions, and can handle data that are represented on different measurement scales. Pruning of DTs can make them smaller and more easily interpretable, while the use of treeboost and tree forest techniques can improve performance.
In this paper, we present several types of decision tree classification algorithms and evaluate them on SPOT5 remote sensing data sets of Conghua area.The decision tree classification algorithms tested include a single decision tree model (CART, CHAID, exaustive CHAID & QUEST) and ensemble decision tree model(TreeBoost&Decsion Tree Forest). Classification accuracies produced by each of these decision tree algorithms are compared with both artificial neural networks and maximum likelihood classifiers. The results showed as follows:
(1)Decision trees in general, and the decision tree forest in particular, produced consistently higher classification accuracies than MLX algorithms. Several factors contribute to this result, the most important being that decision trees can adapt to the noisy and nonlinear relations often observed between land cover classes and remotely sensed data. Decision trees have the further advantage of being nonparametric and therefore make no assumptions regarding the distribution of input data.
(2)Most of tested decision tree algorithms perform better than ANN while some (CHAID & exhaustive CHAID) did not. But neural networks have a number of drawbacks. First, neural networks do not present an easily-understandable model. When looking at decision tree, it is easy to see that some initial variable divides the data into several categories and then other variables split the resulting child groups. This information is very useful to the researcher who
1．在卫星影像的整体分类精度的上，决策树分类技术要优于神经元网络和最大似然分类。相对于神经元网络方法，决策树在训练样本数据的速度要快，并且执行效率要高，对于输入数据空间特征和分类标识，具有更好的弹性和鲁棒性。并且，决策树并不象神经网络方法那样具有黑箱(black box)结构，和分析人员有着良好的交互性和透明性，而神经元网络初始参数的设定是根据实践经验来确定的，缺乏有效的理论指导，带有很强的主观性，并且在训练过程中这些参数要经过不断的调整，才能生成一个较好的网络模型，而且，神经网络的训练非常费时并且对训练结果的好坏事先缺乏判断，这些都是不如决策树分类的地方(在本研究中，虽然神经元网络的分类精度达到了82％，要优于CHAID，但是这种分类精度是在不断调整神经网络的初始参数通过数次尝试才达到的)。相对于最大似然分类，决策树的树状分类结构对数据特征空间分布不需要预先假设某种参数化密度分布，所以其总体分类精度优于传统的参数化统计分类方法。
2．在所有的决策树分类技术中，，组合决策树模型(决策树森林和TreeBoost)的总体表现要比单一决策树模型(CART，CHAID，Exhaustive CHMD，QUEST)模型优秀，但是组合决策树模型的“黑箱”结构其可视性和可解释性又不如单一决策树，正以为如此，在分类过程中要选择那种模型要视情况而定，单一树模型能直观的理解预测变量在分类中的作用和能生成清晰的类别判别规则，而组合树模型通常能比单一树模型得到更高的预测精度。
3．在组合决策树模型中，TreeBoost是通过将每次预测函数的输出赋以一定权重并重复的应用该预测函数来使得总预测误差达到最小化而提高分类精度的。而与TreeBoost模型不同的是，决策树森林中的每棵树都是独立平衡生长的并且它们在所有的树生成之前是不相互影响的，整个森林的预测精度是由其中每一棵树的预测精度组合而得到的。相对于单一树模型，组合树模型能显著提高分类精度，并且该模型能避免过拟合现象，因而不需要对其进行修剪。一般来说，组合树模型树的数目越多，该模型的预测效果越好。
4．在单一决策树模型中，各分类方法的差别主要体现在决策树生长过程中预测变量的选择和变量分割点的选择。在本研究中，CART(精度86.5％)要优于QUEST(82.7％)＞CHAID(81.5％)＞Exhausitve CHAID(81.3％)。
Choice of a classification algorithm is generally based upon a number of factors, among which are availability of software, ease of use, and performance, measured here by overall classification accuracy. The maximum likelihood (ML) procedure is, for many users, the algorithm of choice because of its ready availability and the fact that it does not require an extended training process. Artificial neural networks (ANNs) are now widely used by researchers, but their operational applications are hindered by the need for the user to specify the configuration of the network architecture and to provide values for a number of parameters, both of which affect performance. The ANN also requires an extended training phase.
In the past few years, the use of decision tree (DTs) to classify remotely sensed data has increased. Proponents of the method claim that it has a number of advantages over the ML and ANN algorithms. The DT is computationally fast, make no statistical assumptions, and can handle data that are represented on different measurement scales. Pruning of DTs can make them smaller and more easily interpretable, while the use of treeboost and tree forest techniques can improve performance.
In this paper, we present several types of decision tree classification algorithms and evaluate them on SPOT5 remote sensing data sets of Conghua area.The decision tree classification algorithms tested include a single decision tree model (CART, CHAID, exaustive CHAID & QUEST) and ensemble decision tree model(TreeBoost&Decsion Tree Forest). Classification accuracies produced by each of these decision tree algorithms are compared with both artificial neural networks and maximum likelihood classifiers. The results showed as follows:
(1)Decision trees in general, and the decision tree forest in particular, produced consistently higher classification accuracies than MLX algorithms. Several factors contribute to this result, the most important being that decision trees can adapt to the noisy and nonlinear relations often observed between land cover classes and remotely sensed data. Decision trees have the further advantage of being nonparametric and therefore make no assumptions regarding the distribution of input data.
(2)Most of tested decision tree algorithms perform better than ANN while some (CHAID & exhaustive CHAID) did not. But neural networks have a number of drawbacks. First, neural networks do not present an easily-understandable model. When looking at decision tree, it is easy to see that some initial variable divides the data into several categories and then other variables split the resulting child groups. This information is very useful to the researcher who
引文
[1]李四海．提高遥感数据分类应用的有效途径[J]．国土资源遥感,1995(4)：1-13．
[2]常庆瑞,蒋平安,周勇等．遥感技术导论[M]．北京：科学山版社,2004。
[3] Richard O.Duda, Peter E.Hart, David G.Stork. Pattern Classification [M].Beijing: China Machine Press, 2004.
[4] Morgan J.A. and Sonquist J.N. (1963), Problems in the analysis of survey data: and a proposal, Journal of American Statistical Association, 58, 415-434.
[5] Messenger R.C. and Mandell L.M.(1972), A model search technique for predicative nominal scale multivariate analysis , Journal of American Statistical Association,67,768-772.
[6] Kass G.V. (1980).An Exploratory Technique for Investigating Large Quantities of Categorical Data, 29, No.2, 119-172.
[7] Biggs,D.,B.de Ville, and E.Suen.(1991). A method of choosing multiway partitions for classification and decision trees, Journal of Applied Statistics, 18:49-62.
[8] Hawkins D.M.(1990),FIRM Formal Inference-based Recursive Modeling, Technical Report #546,Department of Applied Statistics, University of Minnesota, St.Paul.
[9] Breiman L., Friedman J.H., Olshen, R.A. and Stone C.J. (1984), Classification and Regression Trees, Wadsworth, Inc.
[10] Loh W.Y. and Shih Y.S.(1997) Split Selection Methods for Classification trees Statistica Sinica 7,815-840.
[11] Loh W.Y. and Vanichsetakul N. (1988) Tree-Structured Classification Via Generalized Discriminant Analysis, Journal of American Statistical Association, 83,715-728.
[12] Loh W.Y.(2002) Regression Trees with Unbiased Variable Selection and Interaction Detection, Statistica Sinica, to appear.
[13] Kim H. and Loh W.Y. (2001) Classification Trees with Unbiased Multiway Splits, Journal of American Statistical Association, and 96,589-604.
[14] Quinlan J.R. (1993) C4.5 Programs for Machine Learning San Mateo, CA, Morgan Kautmann.
[15] Hansen, M. Dubayah, R. and DeFries.R. Classification Trees: An Alternative to Traditional Land Cover Classifiers [J], International Journal of Remote Sensing, 1996, 17(5):1075-1082.
[16] M.A.Friedl and C.E.Brodley (1997).Decision Tree Classification of Land Cover from Remotely Sensed Data [J], Remote Sensing Environment, 61(3):399-409.
[17] Rick L.Lawrence and Andrea Wright. Rule-Based Classification Systems Using Classification and Regression Tree (CART) Analysis [J], PE&RS 2001, 67(10):l 137-1142.
[18] Yoshikawa Masanobu, Shindo Hisakazu and Nishii, Ryuei. Fully automated design of binary decision tree for land cover classification, Proceedings of the 1995 International Geoscience and Remote Sensing Symposium. Part 3 (of 3), Jul 10-14 1995.
[19] Tadjudin Saldju and Landgrebe David A. (1996). Decision tree classifier design for high-dimensional data with limited training samples. Proceedings of the 1996 International Geoscience and Remote Sensing Symposium, IGARSS'96. Part 1 (of 4), May 28-31 1996.
[20] Franklin S. E., Stenhouse, G. B., Hansen, M.J., Popplewell, C.C., Dechka, J. A. and Peddle, D.R.(2001). An Integrated Decision Tree Approach (IDTA) to mapping landcover using satellite remote sensing in support of grizzly bear habitat analysis in the Alberta yellow head ecosystem [J]. Canadian Journal of Remote Sensing, 2001, 27: 579-592.
[21] Simard Marc, Saatchi, Sasan S. and De Grandi, Gianfranco. (2000). Use of decision tree and multiscale texture for classification of JERS-1 SAR data over tropical forest [J]. IEEE Transactions on Geoscience and Remote Sensing, 2000, 38: 2310-2321.
[22] Palaniappan, Zhu Feng, Zhuang Xinhua and Zhao, Yunxin (2000). Enhanced Binary Tree Genetic Algorithm for automatic land cover classification. 2000 International Geoscience and Remote Sensing Symposium (IGARSS 2000), Jul 24-Jul 28 2000, 2: 688-692.
[24] Ding Qiang, Ding Qin and Perrizo (2002). Decision tree classification of spatial data streams using peano count trees. Applied Computing 2002: Proceeedings of the 2002 ACM Symposium on Applied Computing, Mar 11-14 2002, 413-417.
[25] Pal Mahesh and Mather Paul M. (2002). A comparison of decision tree and baekpropagation neural network classifiers for land use classification. 2002 IEEE International Geoseience and Remote Sensing Symposium (IGARSS 2002), Jun 24-28 2002, 503-505.
[26] De Colstoun, Story Michael H. and Thompson Craig. (2002). Vegetation mapping using multi-temporal ETM+ data and a decision tree classifier. 2002 IEEE International Geoscienee and Remote Sensing Symposium (IGARSS 2002), Jun 24-28 2002, 2890-2892.
[27] Saglam, Ersoy Okan and Yazgan, Bingul (2003). Self organizing map and linear support vector machine decision tree with optimized class separability. Smart Engineering System Design: Neural Networks, Fuzzy Logic, Evolutionary Programming, Complex Systems and Artificial Life-Proceedings of the Artificial Neural Networks in Engineering Conference, Nov 2-5 2003, 13: 193-198.
[28] 付炜．土壤遥感分类识别推理决策器的设计[J]，遥感学报，2001，5(6)：434-441。
[29] 韩涛．用TM资料对祁连山部分地区进行针叶林、灌木林分类研究[J]，遥感技术与应用，2002，17(6)：317-321。
[30] Mather, P. M. (1999).Computer processing of remotely-sensed images: An introduction (2nd Ed.). Chichester: Wiley.
[31] Schowengerdt, R. A. (1997).Remote sensing: Models and methods for image procesSing.San Diego: Academic Press.
[32] Foody, G.M., &Arora, M.K. (1997).An evaluation of some factors affecting the accuracy of classification by an artificial neural network. International Journal of Remote Sensing, 18,799-810.
[33] Kavzoglu, T. (2001).An investigation of the design and use of feed-forward artificial neural networks in the classification of remotely sensed images. PhD thesis, University of Nottingham, Nottingham, Nottingham, UK.
[34] Wilkinson, G.G. (1997).Open questions in neurocomputing for earth observation. Neuro-computational in remote sensing data analysis (pp.3-13).Berlin: Springer-Verlag.
[35] Fried, M.A., &Brodley, C.E. (1997).Decision tree classification of land cover from remotely sensed data. Remote Sensing of Environment, 61,399-409.
[36] Gahegan, M., &West, G. (1998).The Classification of complex data sets: An operational comparison of artificial neural networks and decision tree classifiers. Proceedings of 3~(rd) international conference on geocomputation, University of Bristol, UK, 17-19 September 1998, available at http://divcom.otago.ac.nz/SIRC/GeoComp/GeoComp98/61/gc 61.htm. accessed 10 April 2003.
[37] http://www.spotimage.com
[38] Zhang, Yun. (2002). Problems in the fusion of commercial high-resolution satellite as well as Landsat 7 images and initial solutions. In ISPRS, Vol. 34, Part 4, "GeoSpatial Theory, Processing and Applications", Ottawa.
[39] Zhang, Yun. (June 24-28, 2002). A new automatic approach for effectively fusing Landsat 7 as well as IKONOS images. IEEE/IGARSS'02, Toronto, Canada.
[40]田庆久,闵祥军．植被指数研究进展．地球科学进展．1998,13(4)．-327-333．
[41] Jordan C F. (1969).Derivation of leaf area index from quality of light on the forest floor [J].Ecology, 50: 663-666.
[42] Kauth R J, Thomas G S.(1976). The tasseled cap-a graphic description of the spectral-temporal development of agriculture crops as seen by Landsat [A]. Pros Symposium on Machine Processing of Remotely Sensed Data[C]. Purdure University, West Lafayette, Indiana: 41-45.
[43] Wheeler S G, and Misra P N. (1976). Linear dimensionality of landsat agricultural data with implications for classifications [A]. Pros Symposium on Machine Processing of Remotely Sensed Data[C]. West Lafayette, Indiana. Laboratory for the Applications of Remote Sensing.
[44] Jackson, R D, Slater P N, and Printer P J. (1983). Discrimination of growth and water stress in wheat by various vegetation indices through clear and turbid atmospheres.[J].Remote Sens. Environ, 13: 187-208.
[45] Huete A R. (1988). A soil-adjusted vegetation index (SAVI) [J]. Remote Sens. Environ, 25: 295-309.
[46] Elvidge C D, and Z Chen. (1995). Comparison of broad-band and narrow-band red and near-infrared vegetation indices [J]. Remote Sens. Environ, 54: 38-48.
[47] Qi J A. (1994). Modified soil adjusted vegetation index [J]. Remote Sens. Environ, 48: 119-126.
[48] Baret F, Guyot G, Major D J. (1989). TSAVI: A vegetation index which minimizes soil brightness effects on LAI and APAR estimation [A]. Proceedings of the 12th Canadian Symposium on Remote sensing and IGARSS'89[C], Vancouver, Canada, 3: 1355-1358.
[49] Major D J, Baret F, and Guyot G. (1990). A ratio vegetation index adjusted for soil brightness [J]. Int. J. Remote Sens, 11: 727-740.
[50] Kaufman Y J. and Tanre D. (1992). Atmospherically resistant vegetation index (ARVI) for EOS-MODIS [J]. IEEE Trans. on Geosci. And Remote Sensin, 30(2): 261-270.
[51] 张仁华，饶农新，廖国男．(1996)．植被指数的抗大气影响探讨[J]．植物学报，38(1)：53-62
[52] Pinty B, and Verstraete M M. (1992). GEMI: A Non-Linear Index to Monitor Global Vegetation from Satellites [J].Vegetation, 101: 15-20.
[53] Rouse J W, Haas R H, Schell J A, and Deering D W.(1974). Monitoring vegetation systems in the Great Plains with ERTS [A]. Proceedings of Thrid Earth Resources Technology Satellite-1 Symposium[C], Greenbelt: NASA SP-351: 310-317.
[54] 郭妮．植被指数及其研究进展[J]．干旱气象，2003，21(4)：71-75．
[55] Deering D W, Rouse J W, Haas R H, and Schell J A. (1975). Measuring forage production of grazing units from Landsat MSS data [A]. Proceedings of Tenth International Symposium on Remote Sensing of Environment[C], Ann Arbor, ERIM, 2: 1169-1178.
[56] Roujean J L and Breon F M. (1995). Estimating PAR absorbed by vegetation from bidirectional reflectance measurements [J].Remote Sens. Environ, 51: 375-384.
[57] Gitelson A, Kaufman Y J, and Merzlyak M N. (1996). Use of a green channel in remote sensing of global vegetation from EOS-MODIS [J]. Remote Sens. Environ 58(3): 289-298.
[58] 唐世浩，朱启疆，王锦地．(2003)．三波段梯度差值植被指数的理论基础及应用[J]．中国科学(D辑)，33(11)：1094-1102．
[59] Rosenfeld A, Kak A. Digital Picture Processing (2nd edition) M). Academic Press, 1982.
[60] Lee J, Phiipot W. Spectral Textures Pattern Matching: A Classifier for Digital Imagery [J]. IEEE Transactions on Geosciences and Remote Sensing, 1991, 29: 545-548.
[61] Marceau D J, Howaeth P J, Dubois JM, etal. Evaluation of the Gray - level Cooccurrence Matrix Method for L and -cover Classification using SPOT Imagery [J]. IEEE Trans GeosiRemote Sens, 1990, 28(4):513-518.
[62] 杨淑莹，胡军，曹作良．基于图像纹理分析的目标物体识别方法[J]．天津理工学院学报，2001．17(4)：31-33．
[63] 周廷刚，郭志达，盛业华．灰度矢量多波段遥感影像纹理特征及其描述[J]．西安科技学院学报，2000．20(4)：336-340．
[64] Brodley, C. E., &Friedi, M.A. (1996).Identifying and eliminating mislabeled training instances. Proceedings of the 13th national conference on artificial intelligence (pp.799-805).Portland: AAAI Press.
[65] Congalton, R. G. (1991). A review of assessing the accuracy of classification of remotely sensed data. Remote Sensing of Environment, 37, 35-46.
[66] DeFries,R.S.,& Chan,J.C.W.(2000).Multiple criteria for evaluating machine learning algorithms for land cover classification from satellite data. Remote Sensing of Environment, 74,503-515.
[67] DeFries, R.S., Hansen, M., &Townshend, J. (1995). Global discrimination of land cover types from metrics derived from AVHRR pathfinder data. Remote Sensing of Environment, 54,209-222.
[68] Freund, Y., & Schapire, R.E.(1997). A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Sciences, 55,119-139.
[69] Friedl, M.A., Brodley, C.E., & Strahler, A.H.(1999). Maximizing land cover classification accuracies produced by decision trees at continental to global scales. IEEE Transactions on Geoscience and Remote Sensing, GE-37,969-977.
[70] Friedl, M.A., Mclver, D.K., Hodges, J.C.F. & Schaaf, C. (2002).Global land cover mapping from MODIS: algorithms and early results. Remote Sensing of Environment, 83,287-302.
[71] Hansen, M.C., DeFries, R.S., Townshend, J.R.G., &Sohlberg, R. (2000).Global land cover classification at 1km spatial resolution using a classification tree approach. International Journal of Remote Sensing, 21, 1331-1364.
[72] John, G.H. (1995). Robust decision trees: removing outliers from data-bases. Proceedings of the first international conference on knowledge discovery and data mining (pp. 174-179). Menlo Park: AAAI Press.
[73] Quinlan, J.R. (1996). Bagging, boosting and C4.5. Proceedings of the 13~(th) national conference on artificial intelligence (pp.725-730). Portland: AAAI Press.
[74] Quinlan, J.R. (1999). Simplifying decision trees. International Journal of Human-Computer Studies, 51,497-510.
[75] Safavian, S.R., & Landgrebe, D. (1991). A survey of decision tree classifier methodology. IEEE Transactions on Systems, Man, and Cybernetics, 21,660-674.
[76] Ahn, H. and Loh, W.-Y. (1994). Tree-structured proportional hazards regression modeling, Biometrics 50: 471-485.
[77] Breiman, L. (1996b). Bias, variance, and arcing classifiers, Technical Report 460, Department of Statistics, University of California, Berkeley.
[78] Chaudhuri, P., Huang, M.-C., Loh, W.-Y. and Yao, R. (1994). Piecewise-polynomial regression trees, Statistica Sinica 4: 143-167.
[79] Chaudhuri, P., Lo, W.-D., Loh, W.-Y. and Yang, C.-C. (1995). Generalized regression trees, Statistica Sinica 5: 641-666.
[80] Chou, P.A. (1991). Optimal partitioning for classification and regression trees, IEEE Transactions on Pattern Analysis and Machine Intelligence 13:340-354.
[81] Doyle, P. (1973). The use of Automatic Interaction Detector and similar search procedures, Operational Research Quarterly 24: 465-467.
[82] Yan, C. (1995). Regression Trees and Nonlinear Time Series Modeling, PhD thesis. Department of Statistics, University of Wisconsin, Madison.
[83] Friedman, Jerome H. (1999). Stochastic Gradient Boosting. Technical report, Dept. of Statistics, Stanford University.
[84] Friedman, Jerome H. and Bogdan E.Popescu (2003) Importance Sampled Learning Ensembles.
[85] Breiman, Leo ( 2001 ) . "Decision Tree Forests". Machine Learning 45 (1): 5-32, October 2001.
[2]常庆瑞,蒋平安,周勇等．遥感技术导论[M]．北京：科学山版社,2004。
[3] Richard O.Duda, Peter E.Hart, David G.Stork. Pattern Classification [M].Beijing: China Machine Press, 2004.
[4] Morgan J.A. and Sonquist J.N. (1963), Problems in the analysis of survey data: and a proposal, Journal of American Statistical Association, 58, 415-434.
[5] Messenger R.C. and Mandell L.M.(1972), A model search technique for predicative nominal scale multivariate analysis , Journal of American Statistical Association,67,768-772.
[6] Kass G.V. (1980).An Exploratory Technique for Investigating Large Quantities of Categorical Data, 29, No.2, 119-172.
[7] Biggs,D.,B.de Ville, and E.Suen.(1991). A method of choosing multiway partitions for classification and decision trees, Journal of Applied Statistics, 18:49-62.
[8] Hawkins D.M.(1990),FIRM Formal Inference-based Recursive Modeling, Technical Report #546,Department of Applied Statistics, University of Minnesota, St.Paul.
[9] Breiman L., Friedman J.H., Olshen, R.A. and Stone C.J. (1984), Classification and Regression Trees, Wadsworth, Inc.
[10] Loh W.Y. and Shih Y.S.(1997) Split Selection Methods for Classification trees Statistica Sinica 7,815-840.
[11] Loh W.Y. and Vanichsetakul N. (1988) Tree-Structured Classification Via Generalized Discriminant Analysis, Journal of American Statistical Association, 83,715-728.
[12] Loh W.Y.(2002) Regression Trees with Unbiased Variable Selection and Interaction Detection, Statistica Sinica, to appear.
[13] Kim H. and Loh W.Y. (2001) Classification Trees with Unbiased Multiway Splits, Journal of American Statistical Association, and 96,589-604.
[14] Quinlan J.R. (1993) C4.5 Programs for Machine Learning San Mateo, CA, Morgan Kautmann.
[15] Hansen, M. Dubayah, R. and DeFries.R. Classification Trees: An Alternative to Traditional Land Cover Classifiers [J], International Journal of Remote Sensing, 1996, 17(5):1075-1082.
[16] M.A.Friedl and C.E.Brodley (1997).Decision Tree Classification of Land Cover from Remotely Sensed Data [J], Remote Sensing Environment, 61(3):399-409.
[17] Rick L.Lawrence and Andrea Wright. Rule-Based Classification Systems Using Classification and Regression Tree (CART) Analysis [J], PE&RS 2001, 67(10):l 137-1142.
[18] Yoshikawa Masanobu, Shindo Hisakazu and Nishii, Ryuei. Fully automated design of binary decision tree for land cover classification, Proceedings of the 1995 International Geoscience and Remote Sensing Symposium. Part 3 (of 3), Jul 10-14 1995.
[19] Tadjudin Saldju and Landgrebe David A. (1996). Decision tree classifier design for high-dimensional data with limited training samples. Proceedings of the 1996 International Geoscience and Remote Sensing Symposium, IGARSS'96. Part 1 (of 4), May 28-31 1996.
[20] Franklin S. E., Stenhouse, G. B., Hansen, M.J., Popplewell, C.C., Dechka, J. A. and Peddle, D.R.(2001). An Integrated Decision Tree Approach (IDTA) to mapping landcover using satellite remote sensing in support of grizzly bear habitat analysis in the Alberta yellow head ecosystem [J]. Canadian Journal of Remote Sensing, 2001, 27: 579-592.
[21] Simard Marc, Saatchi, Sasan S. and De Grandi, Gianfranco. (2000). Use of decision tree and multiscale texture for classification of JERS-1 SAR data over tropical forest [J]. IEEE Transactions on Geoscience and Remote Sensing, 2000, 38: 2310-2321.
[22] Palaniappan, Zhu Feng, Zhuang Xinhua and Zhao, Yunxin (2000). Enhanced Binary Tree Genetic Algorithm for automatic land cover classification. 2000 International Geoscience and Remote Sensing Symposium (IGARSS 2000), Jul 24-Jul 28 2000, 2: 688-692.
[24] Ding Qiang, Ding Qin and Perrizo (2002). Decision tree classification of spatial data streams using peano count trees. Applied Computing 2002: Proceeedings of the 2002 ACM Symposium on Applied Computing, Mar 11-14 2002, 413-417.
[25] Pal Mahesh and Mather Paul M. (2002). A comparison of decision tree and baekpropagation neural network classifiers for land use classification. 2002 IEEE International Geoseience and Remote Sensing Symposium (IGARSS 2002), Jun 24-28 2002, 503-505.
[26] De Colstoun, Story Michael H. and Thompson Craig. (2002). Vegetation mapping using multi-temporal ETM+ data and a decision tree classifier. 2002 IEEE International Geoscienee and Remote Sensing Symposium (IGARSS 2002), Jun 24-28 2002, 2890-2892.
[27] Saglam, Ersoy Okan and Yazgan, Bingul (2003). Self organizing map and linear support vector machine decision tree with optimized class separability. Smart Engineering System Design: Neural Networks, Fuzzy Logic, Evolutionary Programming, Complex Systems and Artificial Life-Proceedings of the Artificial Neural Networks in Engineering Conference, Nov 2-5 2003, 13: 193-198.
[28] 付炜．土壤遥感分类识别推理决策器的设计[J]，遥感学报，2001，5(6)：434-441。
[29] 韩涛．用TM资料对祁连山部分地区进行针叶林、灌木林分类研究[J]，遥感技术与应用，2002，17(6)：317-321。
[30] Mather, P. M. (1999).Computer processing of remotely-sensed images: An introduction (2nd Ed.). Chichester: Wiley.
[31] Schowengerdt, R. A. (1997).Remote sensing: Models and methods for image procesSing.San Diego: Academic Press.
[32] Foody, G.M., &Arora, M.K. (1997).An evaluation of some factors affecting the accuracy of classification by an artificial neural network. International Journal of Remote Sensing, 18,799-810.
[33] Kavzoglu, T. (2001).An investigation of the design and use of feed-forward artificial neural networks in the classification of remotely sensed images. PhD thesis, University of Nottingham, Nottingham, Nottingham, UK.
[34] Wilkinson, G.G. (1997).Open questions in neurocomputing for earth observation. Neuro-computational in remote sensing data analysis (pp.3-13).Berlin: Springer-Verlag.
[35] Fried, M.A., &Brodley, C.E. (1997).Decision tree classification of land cover from remotely sensed data. Remote Sensing of Environment, 61,399-409.
[36] Gahegan, M., &West, G. (1998).The Classification of complex data sets: An operational comparison of artificial neural networks and decision tree classifiers. Proceedings of 3~(rd) international conference on geocomputation, University of Bristol, UK, 17-19 September 1998, available at http://divcom.otago.ac.nz/SIRC/GeoComp/GeoComp98/61/gc 61.htm. accessed 10 April 2003.
[37] http://www.spotimage.com
[38] Zhang, Yun. (2002). Problems in the fusion of commercial high-resolution satellite as well as Landsat 7 images and initial solutions. In ISPRS, Vol. 34, Part 4, "GeoSpatial Theory, Processing and Applications", Ottawa.
[39] Zhang, Yun. (June 24-28, 2002). A new automatic approach for effectively fusing Landsat 7 as well as IKONOS images. IEEE/IGARSS'02, Toronto, Canada.
[40]田庆久,闵祥军．植被指数研究进展．地球科学进展．1998,13(4)．-327-333．
[41] Jordan C F. (1969).Derivation of leaf area index from quality of light on the forest floor [J].Ecology, 50: 663-666.
[42] Kauth R J, Thomas G S.(1976). The tasseled cap-a graphic description of the spectral-temporal development of agriculture crops as seen by Landsat [A]. Pros Symposium on Machine Processing of Remotely Sensed Data[C]. Purdure University, West Lafayette, Indiana: 41-45.
[43] Wheeler S G, and Misra P N. (1976). Linear dimensionality of landsat agricultural data with implications for classifications [A]. Pros Symposium on Machine Processing of Remotely Sensed Data[C]. West Lafayette, Indiana. Laboratory for the Applications of Remote Sensing.
[44] Jackson, R D, Slater P N, and Printer P J. (1983). Discrimination of growth and water stress in wheat by various vegetation indices through clear and turbid atmospheres.[J].Remote Sens. Environ, 13: 187-208.
[45] Huete A R. (1988). A soil-adjusted vegetation index (SAVI) [J]. Remote Sens. Environ, 25: 295-309.
[46] Elvidge C D, and Z Chen. (1995). Comparison of broad-band and narrow-band red and near-infrared vegetation indices [J]. Remote Sens. Environ, 54: 38-48.
[47] Qi J A. (1994). Modified soil adjusted vegetation index [J]. Remote Sens. Environ, 48: 119-126.
[48] Baret F, Guyot G, Major D J. (1989). TSAVI: A vegetation index which minimizes soil brightness effects on LAI and APAR estimation [A]. Proceedings of the 12th Canadian Symposium on Remote sensing and IGARSS'89[C], Vancouver, Canada, 3: 1355-1358.
[49] Major D J, Baret F, and Guyot G. (1990). A ratio vegetation index adjusted for soil brightness [J]. Int. J. Remote Sens, 11: 727-740.
[50] Kaufman Y J. and Tanre D. (1992). Atmospherically resistant vegetation index (ARVI) for EOS-MODIS [J]. IEEE Trans. on Geosci. And Remote Sensin, 30(2): 261-270.
[51] 张仁华，饶农新，廖国男．(1996)．植被指数的抗大气影响探讨[J]．植物学报，38(1)：53-62
[52] Pinty B, and Verstraete M M. (1992). GEMI: A Non-Linear Index to Monitor Global Vegetation from Satellites [J].Vegetation, 101: 15-20.
[53] Rouse J W, Haas R H, Schell J A, and Deering D W.(1974). Monitoring vegetation systems in the Great Plains with ERTS [A]. Proceedings of Thrid Earth Resources Technology Satellite-1 Symposium[C], Greenbelt: NASA SP-351: 310-317.
[54] 郭妮．植被指数及其研究进展[J]．干旱气象，2003，21(4)：71-75．
[55] Deering D W, Rouse J W, Haas R H, and Schell J A. (1975). Measuring forage production of grazing units from Landsat MSS data [A]. Proceedings of Tenth International Symposium on Remote Sensing of Environment[C], Ann Arbor, ERIM, 2: 1169-1178.
[56] Roujean J L and Breon F M. (1995). Estimating PAR absorbed by vegetation from bidirectional reflectance measurements [J].Remote Sens. Environ, 51: 375-384.
[57] Gitelson A, Kaufman Y J, and Merzlyak M N. (1996). Use of a green channel in remote sensing of global vegetation from EOS-MODIS [J]. Remote Sens. Environ 58(3): 289-298.
[58] 唐世浩，朱启疆，王锦地．(2003)．三波段梯度差值植被指数的理论基础及应用[J]．中国科学(D辑)，33(11)：1094-1102．
[59] Rosenfeld A, Kak A. Digital Picture Processing (2nd edition) M). Academic Press, 1982.
[60] Lee J, Phiipot W. Spectral Textures Pattern Matching: A Classifier for Digital Imagery [J]. IEEE Transactions on Geosciences and Remote Sensing, 1991, 29: 545-548.
[61] Marceau D J, Howaeth P J, Dubois JM, etal. Evaluation of the Gray - level Cooccurrence Matrix Method for L and -cover Classification using SPOT Imagery [J]. IEEE Trans GeosiRemote Sens, 1990, 28(4):513-518.
[62] 杨淑莹，胡军，曹作良．基于图像纹理分析的目标物体识别方法[J]．天津理工学院学报，2001．17(4)：31-33．
[63] 周廷刚，郭志达，盛业华．灰度矢量多波段遥感影像纹理特征及其描述[J]．西安科技学院学报，2000．20(4)：336-340．
[64] Brodley, C. E., &Friedi, M.A. (1996).Identifying and eliminating mislabeled training instances. Proceedings of the 13th national conference on artificial intelligence (pp.799-805).Portland: AAAI Press.
[65] Congalton, R. G. (1991). A review of assessing the accuracy of classification of remotely sensed data. Remote Sensing of Environment, 37, 35-46.
[66] DeFries,R.S.,& Chan,J.C.W.(2000).Multiple criteria for evaluating machine learning algorithms for land cover classification from satellite data. Remote Sensing of Environment, 74,503-515.
[67] DeFries, R.S., Hansen, M., &Townshend, J. (1995). Global discrimination of land cover types from metrics derived from AVHRR pathfinder data. Remote Sensing of Environment, 54,209-222.
[68] Freund, Y., & Schapire, R.E.(1997). A decision-theoretic generalization of on-line learning and an application to boosting. Journal of Computer and System Sciences, 55,119-139.
[69] Friedl, M.A., Brodley, C.E., & Strahler, A.H.(1999). Maximizing land cover classification accuracies produced by decision trees at continental to global scales. IEEE Transactions on Geoscience and Remote Sensing, GE-37,969-977.
[70] Friedl, M.A., Mclver, D.K., Hodges, J.C.F. & Schaaf, C. (2002).Global land cover mapping from MODIS: algorithms and early results. Remote Sensing of Environment, 83,287-302.
[71] Hansen, M.C., DeFries, R.S., Townshend, J.R.G., &Sohlberg, R. (2000).Global land cover classification at 1km spatial resolution using a classification tree approach. International Journal of Remote Sensing, 21, 1331-1364.
[72] John, G.H. (1995). Robust decision trees: removing outliers from data-bases. Proceedings of the first international conference on knowledge discovery and data mining (pp. 174-179). Menlo Park: AAAI Press.
[73] Quinlan, J.R. (1996). Bagging, boosting and C4.5. Proceedings of the 13~(th) national conference on artificial intelligence (pp.725-730). Portland: AAAI Press.
[74] Quinlan, J.R. (1999). Simplifying decision trees. International Journal of Human-Computer Studies, 51,497-510.
[75] Safavian, S.R., & Landgrebe, D. (1991). A survey of decision tree classifier methodology. IEEE Transactions on Systems, Man, and Cybernetics, 21,660-674.
[76] Ahn, H. and Loh, W.-Y. (1994). Tree-structured proportional hazards regression modeling, Biometrics 50: 471-485.
[77] Breiman, L. (1996b). Bias, variance, and arcing classifiers, Technical Report 460, Department of Statistics, University of California, Berkeley.
[78] Chaudhuri, P., Huang, M.-C., Loh, W.-Y. and Yao, R. (1994). Piecewise-polynomial regression trees, Statistica Sinica 4: 143-167.
[79] Chaudhuri, P., Lo, W.-D., Loh, W.-Y. and Yang, C.-C. (1995). Generalized regression trees, Statistica Sinica 5: 641-666.
[80] Chou, P.A. (1991). Optimal partitioning for classification and regression trees, IEEE Transactions on Pattern Analysis and Machine Intelligence 13:340-354.
[81] Doyle, P. (1973). The use of Automatic Interaction Detector and similar search procedures, Operational Research Quarterly 24: 465-467.
[82] Yan, C. (1995). Regression Trees and Nonlinear Time Series Modeling, PhD thesis. Department of Statistics, University of Wisconsin, Madison.
[83] Friedman, Jerome H. (1999). Stochastic Gradient Boosting. Technical report, Dept. of Statistics, Stanford University.
[84] Friedman, Jerome H. and Bogdan E.Popescu (2003) Importance Sampled Learning Ensembles.
[85] Breiman, Leo ( 2001 ) . "Decision Tree Forests". Machine Learning 45 (1): 5-32, October 2001.