第五章 結論與建議
第二節 建議
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
第二節 建議
根據本研究之結論，針對適用於多影像的崩塌地偵測機制的建立，有 以下幾點建議：
1. 於本研究，產出各類別訓練物件的分割尺度由試誤法決定，而作 業流程中的分割策略(陰影區的尺度參數初始值小於非陰影區─
陰影區與非陰影區的劃分)又基於訓練資料；且本研究的尺度參 數設定僅基於一張影像的測試成果，然各影像的「物件明亮度分 布曲線」(即使位於明亮度門檻值之同側)型態各異，一張影像的 測試成果不具代表性。建議深入探討光譜異質性權重、陰影與尺 度參數三者的關係。
2. 網格搜索成果的品質受自訂的搜尋值域與網格間距所影響，未必 為最佳參數，可嘗試其他參數優化方法。
3. 本研究所使用的 9 公尺空間解析度 DEM 僅提供粗略的地形資 訊，未達地形學分析的精度需求。建議使用較精細的光達地形資 料(1 公尺空間解析度)。
4. 未訓練類別的誤判物徵無法由地形特徵過濾準則除盡。建議增加 訓練類別、於第一階段進行初步辨識，或於第二階段納入現有的 輔助資料，如河川向量圖、地物高度模型(OHM)、道路圖…等。
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
‧
‧
葉雅馨，2006，「用 LiDAR 與航照影像於崩塌地判釋之研究」，國立嘉 義大學土木與水資源工程研究所碩士論文：嘉義。
‧
Optimization Approach for High Quality Multi-scale Image Segmentation.”pp. 12-23 in Angewandte Geographische Informationsverarbeitung XII, edited by Strobl J., Karlsruhe.
eCognition, 2011, eCognition Developer Reference Book. Trimble Germany GmbH, Trappentreustr., München, Germany.
eCognition, 2011, eCognition Developer User Guide. Trimble Germany GmbH, Trappentreustr., München, Germany.
Hall, M., 1999, “Correlation-based Feature Selection for Machine Learning”, Unpublished doctoral dissertation, Department of Computer Science, University of Waikato, Hamilton, New Zealand.
Harrington, P., 2012, Machine Learning in Action, Manning.
Horn, B.K.P., 1981, “Hill shading and the reflectance map”, Proceedings of IEEE, 69(1): 14-47.
J. Richard Landis and Gary G. Koch, “The Measurement of Observer
Agreement for Categorical Data”, Biometrics, Vol. 33, No. 1 (Mar., 1977), pp. 159-174
Jensen, J.R., 2006, Introductory Digital Image Processing, Prentice Hall.
Kim, M. and Madden, M., 2006, “Determination of Optimal Scale Parameter for Alliance-level Forest Classification of Multispectral IKONOS Images”,
‧
Paper presented at the International Society for Photogrammetry and Remote Sensing, Dresden, Germany.
Liang, S., Fang, H., and Chen, M., 2001, “Atmospheric Correction of Landsat ETM+ Land Surface Imagery-Part I: Methods”, Geoscience and Remote
Sensing, 39(11): 2490-2498.
Martha, T., Kerle, N., Jetten, V., Westen, C., and Kumar, K., 2009,
“Characterising Spectral, Spatial and Morphometric Properties of Landslides for Semi-automatic Detection Using Object-oriented Methods”,
Geomorphology, 116(2010): 24-36.
Mayr, W. and Heipke, C., 1988, “A Contribution to Digital Orthophoto Generation”, Paper presented at the International Archives of Photogrammetry and Remote Sensing, Kyoto, Japan.
Pekkarinen, A., 2002, “A Method for the Segmentation of Very High Resolution Images of Forested Landscapes”, International Journal of
Remote Sensing, 23(14): 2817-2836.
Rizvi, I. and Mohan, B., 2010, “Improving the Accuracy of Object Based Supervised Image Classification using Cloud Basis Function Neural Network for High Resolution Satellite Images”, International Journal of
Image Processing (IJIP), 4(4): 342-353.
Song, M. and Civco, D.L., 2002, “A Knowledge-based Approach for Reducing Cloud and Shadow”, Paper presented at the ASPRS-ACSM Annual Conference and FIG XXII Congress, Washington, DC, USA, April 19-April 26.
Stumpf, A. and Kerle, N., 2010, “Object-oriented mapping of landslides using Random Forests”, Remote Sensing of Environment, 115(2011): 2564-2577.
‧
Pixel-based Classification for Mapping Mangroves with IKONOS Imagery”, International Journal of Remote Sensing, 25(24): 5655-5668.Xiao, P., Feng, X., An, R., and Zhao, S., 2010, “Segmentation of Multispectral High-resolution Satellite Imagery Using Log Gabor Filters”, International
Journal of Remote Sensing, 31(6): 1427-1439.
Zhou, W., Huang, G., Troy, A., and Cadenasso M.L., 2009, “Object-based Land Cover Classification of Shaded Areas in High Spatial Resolution Imagery of Urban Areas: A Comparison Study”, Remote Sensing of Environment, 113: 1769-1777.
三、網頁部分
行 政 院 農 業 委 員 會 林 務 局 ， 2010 ， 航 空 攝 影 紀 錄 資 料 查 詢 。 http://www.afasi.gov.tw/，取用日期：2014 年 03 月 05 日。
瑞竣科技，2011，eCognition。http://www.richitech.com.tw/，取用日期：
PostgreSQL (2013). About. Retrieved January 2, 2014 from PostgreSQL on the World Wide Web: http://www.postgresql.org/.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Pyshp - Python Shapefile Library (2013). Project Home. Retrieved January 10, 2014 from Pyshp on the World Wide Web:
http://code.google.com/p/pyshp/.
PyWEKA (2014). API. Retrieved June 8, 2014 from PyWEKA on the World Wide Web: http://pythonhosted.org/python-weka-wrapper/api.html.
scikit learn (2013). User Guide. Retrieved April 25, 2014 from scikit learn on the World Wide Web: http://scikit-learn.org/stable/user_guide.html.
Wikipedia (2014). Probability density function. Retrieved June 24, 2014 from Wikipedia on the World Wide Web:
http://en.wikipedia.org/wiki/Probability_density_function.