臺灣博碩士論文加值系統

　　遙測影像土地利用(Land-use/land-cover, LULC)分類被廣泛應用於地球表面監測、環境變遷偵測與水資源管理等領域。分類後常以分類主題圖呈現遙測影像分類的結果，並提供分類主題圖之分類正確率。遙測影像分類中監督式分類法需以訓練資料建立分類規則。由於訓練資料為從遙測影像選取的樣本資料，訓練資料的選取會造成分類正確率的不確定性。傳統上多數研究評估分類正確率時，會從遙測影像中獨立選取一組參考資料，對其分類建立混淆矩陣(confusion matrix)計算分類正確率。然而採用不同的參考資料可能求得顯著不同的分類正確率，且此分類正確率僅能代表以所選訓練資料建立之分類規則的分類效果，因此以一組參考資料混淆矩陣難以有效評估分類主題圖之分類正確率。此外，混淆矩陣呈現的分類正確率也會受到被分類資料的各類別抽樣數量影響。
　　以混淆矩陣評估分類效果的優勢在於可以呈現每個類別之分類情形，然而受到樣本選擇的影響，混淆矩陣呈現的分類正確率具有不確定性。本研究以多光譜遙測影像之序率模擬探討遙測影像土地利用之分類正確率評估，包括類別間樣本數比例對混淆矩陣的影響、訓練資料與參考資料選取造成的分類不確定性，與量化分類不確定性的方法。研究結果顯示(一)類別間樣本數比例對於分類正確率的評估具有決定性的影響 (二)訓練資料與參考資料的選擇會影響分類正確率的評估，而拔靴法訓練資料分類正確率可以做為全幅影像分類正確率(global accuracy)的估計(三)以一組訓練資料求得之拔靴法(Bootstrap)信賴區間，可做為全幅影像分類正確率的95%信賴區間。
　　對於遙測影像土地利用分類之不確定性，本研究總結以混淆矩陣評估分類正確率時，各類別樣本佔所有資料比例應與遙測影像中各類別被觀察到的機率相同。監督式分類的分類正確率評估應考慮訓練資料選取造成的的不確定性，為量化分類主題圖中各類別的分類不確定性，可以一組訓練資料透過拔靴法重覆取樣，對各類別使用者、生產者分類正確率與總體分類正確率分別求得全幅影像分類正確率的95%信賴區間，以此方式描述全幅影像分類結果之正確率，可以有效控制估計錯誤的機會。

Land-use/land-cover (LULC) classification using remote sensing images have been widely applied for earth surface monitoring, environmental change detection, water resources management, etc. All supervised classification methods require using a set of training data to establish class-assignment rules for pixels of unknown classes. A confusion matrix (or error matrix) which summarizes classification results of the training data or an independent set of reference data is then used to assess the classification accuracies of individual classes. However, classification accuracies of the training or reference data presented in the confusion matrix are estimates of the true and unknown classification accuracies of the population, i.e. all pixels of individual LULC classes. The class-assignment rules are derived from training data which are samples of individual classes, thus classification accuracies are inherently associated with uncertainties, i.e. sample variability would cause the uncertainty of classification accuracy. Besides, the ratio between sample sizes of difference classes could affect the classification accuracy. Conventional way of evaluating the LULC classification results is to choose an independent set of reference data and use the decision rules established by the training dataset for reference data LULC classification, and then examine the reference-data-based confusion matrix. We argue that such practices have limited capability in providing meaningful assessments since different reference datasets can yield significantly different accuracies in the reference-data-based confusion matrix. In this study, we investigated the uncertainty in classification accuracies (including the producer’s, user’s and overall accuracies) by using a bootstrap resampling approach. Unlike most LULC classifications, the proposed approach can provide 95% confidence intervals for class-specific classification accuracies. We also showed the theoretical basis and demonstrated that the conventional reference-data-based confusion matrix has very limited capability in assessing LULC classification results. Therefore, we conclude that LULC classification results should be evaluated by assessing the uncertainty of training-data-based confusion matrix and confidence intervals of class-specific classification accuracies can be provided by using the proposed bootstrap resampling approach.

摘要 i
ABSTRACT ii
目錄 iii
表目錄 v
圖目錄 vi
第一章 前言 1
1.1 研究動機與目的 1
1.2 研究架構及流程 2
第二章 文獻回顧 5
2.1 遙測影像分類 5
2.2 分類正確率的評估 7
2.3 抽樣方法 8
2.4 分類正確率的不確定性 12
第三章 理論與方法 15
3.1 遙測影像分類 15
3.1.1 多光譜遙測影像 15
3.1.2 高斯最大概似分類法 16
3.1.3 貝氏分類法 18
3.1.4 多變量常態分配參數推估 19
3.1.5 混淆矩陣 20
3.1.6 拔靴法 23
3.2 分類正確率之評估 26
3.2.1 傳統分類正確率之評估 26
3.2.2 定義各分類正確率表示方式 26
3.2.3 模擬方法與流程 27
第四章 研究資料與方法 33
4.1 遙測影像模擬資料 33
4.2 特徵空間與全幅影像分類正確率 35
第五章 結果與討論 39
5.1 分類規則的建置 39
5.1.1 訓練資料之樣本大小 39
5.1.2 雙變量常態分配參數推估 40
5.2 類別間樣本數比例對於分類結果的影響 41
5.3 參考資料混淆矩陣之不確定性 51
5.4 訓練資料混淆矩陣之不確定性 56
5.5 綜合分析全幅影像分類正確率之估計方式 59
5.6 以拔靴法評估各類別之分類不確定性 63
5.7 全幅影像分類正確率的信賴區間 71
第六章 結論 73
參考文獻 75

周心怡。拔靴法(Bootstrap)之探討及其應用。碩士論文，國立中央大學桃園市，2004。
潘國樑。遙測學大綱-遙測概念,原理與影像判釋技術，第二版。科技圖書股份有限公司，台北市，2009。
Aronoff, S. Remote sensing for GIS managers; ESRI Press: Redlands, California, USA, 2005.
Bickel, P. J.; Doksum, K. A. Mathematical statistics: basic ideas and selected topics, volume I, 2nd ed.; CRC Press, 2015.
Champagne, C.; McNairn, H.; Daneshfar, B.; Shang, J. A bootstrap method for assessing classification accuracy and confidence for agricultural land use mapping in Canada. International Journal of Applied Earth Observation and Geoinformation 2014, 29, 44-52.
Chapin Iii, F. S.; Zavaleta, E. S.; Eviner, V. T.; Naylor, R. L.; Vitousek, P. M.; et al. Consequences of changing biodiversity. Nature 2000, 405(6783), 234-242.
Chen, Y. C.; Chiu, H. W.; Su, Y. F.; Wu, Y. C.; Cheng, K. S. Does urbanization increase diurnal land surface temperature variation? Evidence and implications. Landscape and Urban Planning 2017, 157, 247-258.
Cochran, W. G. Sampling techniques, 2nd ed.; John Wiley and Sons: New York, USA, 1963.
Congalton, R. G. A review of assessing the accuracy of classifications of remotely sensed data. Remote Sensing of Environment 1991, 37(1), 35-46.
Congalton, R. G.; Green, K. A practical look at the sources of confusion in error matrix generation. Photogrammetric Engineering and Remote Sensing 1993, 59(5), 641-644.
Congalton, R. G.; Green, K. Assessing the accuracy of remotely sensed data: principles and applications; Lewis Publishers: Boca Raton, Florida, USA, 1999.
Congalton, R. G.; Plourde, L. Quality assurance and accuracy assessment of information derived from remotely sensed data. In Manual of geospatial science and technology; Bossler, J. D., Jensen, J. R., McMaster, R. B., Rizos C., Eds.; CRC Press, 2002; pp. 349-361.
Czaplewski, R. L. Variance approximations for assessments of classification accuracy. Research Paper, Res. Pap. RM-RP-316. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Forest and Range Experiment Station. 29 p, 1994.
Demirkesen, A. C.; Evrendilek, F.; Berberoglu, S.; Kilic, S. Coastal flood risk analysis using Landsat-7 ETM+ imagery and SRTM DEM: A case study of Izmir, Turkey. Environmental monitoring and assessment 2007, 131(1-3), 293-300.
Di Stefano, J. A confidence interval approach to data analysis. Forest Ecology and Management 2004, 187(2), 173-183.
Douglas, I. Hydrological investigations of forest disturbance and land cover impacts in South–East Asia: a review. Philosophical Transactions of the Royal Society of London B: Biological Sciences 1999, 354(1391), 1725-1738.
Efron, B. Bootstrap methods: Another look at the jackknife. The Annals of Statistics 1979, 7(1), 1-26.
Efron, B.; Tibshirani, R. J. An introduction to the bootstrap.; CRC press, 1994.
Fitzgerald, R.; Lees, B. Assessing the classification accuracy of multisource remote sensing data. Remote Sensing of Environment 1994, 47(3), 362-368.
Fitzpatrick-Lins, K. Comparison of sampling procedures and data analysis for a land-use and land-cover map. Photogrammetric Engineering and Remote Sensing 1981, 47(3), 343-351.
Fleiss, J. L.; Levin, B.; Paik, M. C. Statistical methods for rates and proportions. John Wiley and Sons: New York, USA, 2013.
Foody, G. M. Status of land cover classification accuracy assessment. Remote Sensing of Environment 2002, 80(1), 185-201.
Foody, G. M. Sample size determination for image classification accuracy assessment and comparison. International Journal of Remote Sensing 2009, 30(20), 5273-5291.
Franklin, S. E.; Peddle, D. R.; Wilson, B. A.; Blodgett, C. F. Pixel sampling of remotely sensed digital imagery. Computers & Geosciences 1991, 17(6), 759-775.
Gopal, S.; Woodcock, C. Theory and methods for accuracy assessment of thematic maps using fuzzy sets. Photogrammetric Engineering and Remote Sensing 1994, 60(2), 181-188.
Hess, G. R.; Bay, J. M. Generating confidence intervals for composition-based landscape indexes. Landscape Ecology 1997, 12(5), 309-320.
Hsiao, L. H.; Cheng, K. S. Assessing uncertainty in LULC classification accuracy by using bootstrap resampling. Remote Sensing 2016, 8(9), 705.
Hung, W. C.; Chen, Y. C.; Cheng, K. S. Comparing landcover patterns in Tokyo, Kyoto, and Taipei using ALOS multispectral images. Landscape and Urban Planning 2010, 97(2), 132-145.
Janssen, L. L. F.; van der Wel, F. J. M. Accuracy assessment of satellite derived landcover data: A review. Photogrammetric engineering and remote sensing 1994, 60(4), 419-426.
Johnson R. A. and Wichern D. W. Applied multivariate statistical analysis, 6th ed.; Prentice Hall: Upper Saddle River, NJ, USA, 2007; pp. 170-171.
Kohavi, R. A study of cross-validation and bootstrap for accuracy estimation and model selection. In C. S. Mellish, ed., Proceedings of the 14th International Joint Conference on Articial Intelligence; Morgan Kaufmann: Montreal, Canada. San Mateo, CA, 1995; pp. 1137-1143.
Lang, R.; Shao, G.; Pijanowski, B. C.; Farnsworth, R. L. Optimizing unsupervised classifications of remotely sensed imagery with a data-assisted labeling approach. Computers & Geosciences 2008, 34(12), 1877-1885.
Lillesand, T.; Kiefer, R. W.; Chipman, J. Remote sensing and image interpretation, 7nd ed.; John Wiley and Sons: New York, USA, 2015.
Lu, D.;Weng, Q. A survey of image classification methods and techniques for improving classification performance. International Journal of Remote Sensing 2007, 28(5), 823-870.
Lunetta, R. S.; Lyon, J. G. Remote sensing and GIS accuracy assessment. CRC press, 2004.
McCaffrey, T. M.; Franklin, S. E. Automated training site selection for large-area remote-sensing image analysis. Computers & Geosciences 1993, 19(10), 1413-1428.
Muller, S.; Walker, D.; Nelson, F.; Auerback, N.; Bockheim, J.; et al. Accuracy assessment of a land-cover map of the Kuparuk river basin, Alaska: considerations for remote regions. Photogrammetric Engineering and Remote Sensing 1998, 64(6), 619-628.
Pal, M. Random forest classifier for remote sensing classification. International Journal of Remote Sensing 2005, 26(1), 217-222.
Penner, J. E. Atmospheric chemistry and air quality. In Changes in land use and land cover: a global perspective; Meyer, W. B., Turner II B. L., Eds.; ‪Cambridge University Press: New York, USA ,‬ 1994; pp. 175-209.‬‬‬‬‬‬‬‬‬‬‬‬‬
Powell, R.; Matzke, N.; De Souza, C.; Clark, M.; Numata, I.; et al. Sources of error in accuracy assessment of thematic land-cover maps in the Brazilian Amazon. Remote Sensing of Environment 2004, 90(2), 221-234.
Saeed, N.; Pervaiz, M. K.; Shahbaz, M. Q. Determination of sample size. European Journal of Scientific Research 2005, 14(3), 319-325.
Stehman, S. Thematic map accuracy assessment from the perspective of finite population sampling. Remote Sensing 1995, 16(3), 589-593.
Stehman, S. V. Design, analysis, and inference for studies comparing thematic accuracy of classified remotely sensed data: a special case of map comparison. Journal of Geographical Systems 2006, 8(2), 209-226.
Stehman, S. V.; Czaplewski, R. L. Design and analysis for thematic map accuracy assessment: fundamental principles. Remote Sensing of Environment 1998, 64(3), 331-344.
Story, M.; Congalton, R. G. Accuracy assessment: a user’s perspective. Photogrammetric Engineering and Remote Sensing 1986, 52(3), 397-399.
Tortora, R. D. A note on sample size estimation for multinomial populations. The American Statistician 1978, 32(3), 100-102.
Vitousek, P. M. Beyond global warming: ecology and global change. Ecology 1994, 75(7), 1861-1876.
Xu, H. Modification of normalised difference water index (NDWI) to enhance open water features in remotely sensed imagery. International journal of remote sensing 2006, 27(14), 3025-3033.