臺灣山地面積約占全島面積70%，在多數河川位於山區情況下，河床地形數據難以頻繁調查並更新，而若以錯誤之河川斷面數據進行流域治理或河川工程規劃設計，將可能導致設計失敗並進一步影響保全標的之生命財產安全。考量現有河床地形調查技術之侷限性，本研究以搭載多光譜相機之無人飛行載具，以綠光光譜、紅光光譜、近紅外光光譜、常態化差異植被指數(Normalized Difference Vegetation Index, NDWI)、常態化差異水體指數(Normalized Difference Water Index, NDWI)等因子，結合傳統線性之迴歸分析(Regression Analysis, REG)與人工智慧演算法中之基因表達程序編輯(Gene-expression Programming, GEP)以及類神經網路(Artificial Neural Network, ANN)，針對台灣山區河道特性，開發可應用於淺水河段之水深模擬模式，並以精度評估指標，包括判定係數(Coefficient of Determination, R2)、均方根誤差(Root Mean Squared Error, RMSE)、絕對誤差率(Absolute Error Rate, AER)以及平均絕對誤差(Mean Absolute Error, MAE)，並以敏感度分析(Sensitivity Analysis, SA)，評估各因子之間對水深模擬之相對重要性。
在模型精度評估部分，REG、ANN和GEP之R2分別為0.7294、0.7916和0.801，MAE分別為0.1845、0.1566和0.1544 m，RMSE分別為0.2249、0.1996和0.1950 m。與REG模型相比，GEP和ANN模型都表現出較低的MAE和RMSE。與前人研究結果相比，本研究發展之模型具有相對較低之MAE和RMSE。在最佳模式評估方面，比較3個模式在模型測試階段之AER，並以AER < 20%視為高精度，成果顯示GEP及ANN之高精度比例分別為51%及58%，而REG僅有39%，因此建議在未來應用中以ANN進行水深模擬。最後在ANN之SA評估中，發現NIR和NDWI為最高影響因素，其中NIR與水深成反比，而NDWI與水深成正比關係。
整體而言，本研究開發之模式符合主管機關對於水道橫斷面測量規範中，以間接方式藉由水深反算水下斷面之標準，即河槽部分容許誤差≤ 25 cm之規定，對台灣山區淺水河段之地形測量、河流管理和災害預防等實務工作具有相當程度之助益。一般商用型之無人飛行載具，若引入本文之設定及分析模式，可於拍攝條件許可下即時取得資料。相對於衛星及有人機載具，本文建議使用之多光譜影像技術，取得山區淺水河段之水深的方法，更具有低風險、低成本、高效率與高空間解析度之優點。然於實際應用中，仍需考慮模型之發展基礎、精度要求和現場調查環境以選擇合適之模型。在未來研究中，建議擴大調查範圍或以薈萃分析(Meta-analysis)，結合新型之AI深度學習技術，進一步提高河床地形及斷面調查技術之準確性和合理性。

River bed topography surveys are essential for various applications but are challenging in Taiwan's mountainous regions due to frequent rainfall and terrain difficulties. This study addresses the difficulty of mountainous river bed topography surveys by introducing an innovative approach. A multispectral camera-equipped unmanned aerial vehicle (UAV) was used, coupled with conventional simple linear regression (REG) and artificial intelligence (AI) algorithms, namely artificial neural network (ANN) and gene-expression programming (GEP). After integrating the obtained images with 171 water depth measurements for bathymetry modeling, it was found that the AI-based models outperformed the REG model in terms of accuracy. The results show that the coefficient of determination (R2) of REG, ANN and GEP is 0.7294, 0.7916 and 0.801, the mean absolute error (MAE) is 0.1845, 0.1566 and 0.1544 m, and root mean square error (RMSE) is 0.2249, 0.1996 and 0.1950 m. Spectral reflection and sensitivity analysis results reveal that the relationship between near-infrared (NIR) reflection and water depth significantly decreases when the water depth is less than 0.8 m, indicating the potential for greater error in bathymetry simulation using NIR in low water depths. Overall, compared with satellites and manned aircraft vehicles, UAV has the advantages of low risk, low cost, high efficiency and high spatial resolution, and can obtain data in real time when shooting conditions available. Their versatility implies potential use in environments such as shallow rivers or near-shore areas under similar conditions. In practical applications, appropriate model selection should still be considered, along with model development foundation, precision requirements, and on-site survey situation.

Alpers, W., & Hennings, I. A theory of the imaging mechanism of underwater bottom topography by real and synthetic aperture radar. 89(C6), 10529-10546. doi:https://doi.org/10.1029/JC089iC06p10529 (1984).
Ashphaq, M., Srivastava, P. K., & Mitra, D. Review of near-shore satellite derived bathymetry: Classification and account of five decades of coastal bathymetry research. Journal of Ocean Engineering and Science, 6(4), 340-359. doi:https://doi.org/10.1016/j.joes.2021.02.006 (2021).
Bangen, S. G., Wheaton, J. M., Bouwes, N., Bouwes, B., & Jordan, C. A methodological intercomparison of topographic survey techniques for characterizing wadeable streams and rivers. Geomorphology, 206, 343-361. doi:https://doi.org/10.1016/j.geomorph.2013.10.010 (2014).
Beget, M. E., & Di Bella, C. M. Flooding: The effect of water depth on the spectral response of grass canopies. Journal of Hydrology, 335(3), 285-294. doi:https://doi.org/10.1016/j.jhydrol.2006.11.018 (2007).
Benny, A. H., & Dawson, G. J. Satellite Imagery as an Aid to Bathymetric Charting in the Red Sea. The Cartographic Journal, 20(1), 5-16. doi:10.1179/caj.1983.20.1.5 (1983).
Bierwirth, P., Lee, T., Burne, R. J. P. E., & Sensing, R. Shallow sea-floor reflectance and water depth derived by unmixing multispectral imagery. 59(3). (1993).
Bishop, C. M., & Nasrabadi, N. M. Pattern recognition and machine learning (Vol. 4): Springer.(2006).
Bures, L., Sychova, P., Maca, P., Roub, R., & Marval, S. River Bathymetry Model Based on Floodplain Topography. Water, 11(6). doi:10.3390/w11061287 (2019).
Calkoen, C. J., Hesselmans, G. H. F. M., Wensink, G. J., & Vogelzang, J. The Bathymetry Assessment System: Efficient depth mapping in shallow seas using radar images. International Journal of Remote Sensing, 22(15), 2973-2998. doi:10.1080/01431160116928 (2001).
Campbell, D. E., & Spinrad, R. W. The relationship between light attenuation and particle characteristics in a turbid estuary. Estuarine, Coastal and Shelf Science, 25(1), 53-65. doi:https://doi.org/10.1016/0272-7714(87)90025-4 (1987).
Carlson, T. N., & Ripley, D. A. On the relation between NDVI, fractional vegetation cover, and leaf area index. Remote Sensing of Environment, 62(3), 241-252. doi:https://doi.org/10.1016/S0034-4257(97)00104-1 (1997).
Dadson, S. J., Hovius, N., Chen, H., Dade, W. B., Hsieh, M.-L., Willett, S. D., . . . Lin, J.-C. Links between erosion, runoff variability and seismicity in the Taiwan orogen. Nature, 426(6967), 648-651. doi:10.1038/nature02150 (2003).
Davis, S. M., Landgrebe, D. A., Phillips, T. L., Swain, P. H., Hoffer, R. M., Lindenlaub, J. C., & Silva, L. F. J. N. Y. Remote sensing: the quantitative approach. (1978).
De Loor, G. P., & Brunsveld van Hulten, H. W. Microwave measurements over the North Sea. Boundary-Layer Meteorology, 13(1), 119-131. doi:10.1007/BF00913866 (1978).
DJI. P4 Multispectral. Retrieved from https://www.dji.com/tw/p4-multispectral. (2023).
Ferreira, C. Gene Expression Programming in Problem Solving. In R. Roy, M. Köppen, S. Ovaska, T. Furuhashi, & F. Hoffmann (Eds.), Soft Computing and Industry: Recent Applications (pp. 635-653). London: Springer London. (2002).
Ferreira, F. L., Bota, D. P., Bross, A., Mélot, C., & Vincent, J.-L. J. J. Serial evaluation of the SOFA score to predict outcome in critically ill patients. 286(14), 1754-1758. (2001).
Fu, L., & Holt, B. SEASAT views oceans and sea ice with Synthetic Aperture Radar. JPL Publ., 81–120. (1982).
Guenther, G. C., Brooks, M. W., & LaRocque, P. E. J. R. S. o. E. New capabilities of the “SHOALS” airborne lidar bathymeter. 73(2), 247-255. (2000).
Heblinski, J., Schmieder, K., Heege, T., Agyemang, T. K., Sayadyan, H., & Vardanyan, L. High-resolution satellite remote sensing of littoral vegetation of Lake Sevan (Armenia) as a basis for monitoring and assessment. Hydrobiologia, 661(1), 97-111. doi:10.1007/s10750-010-0466-6 (2011).
Hennings, I. Radar imaging of submarine sand waves in tidal channels. 95(C6), 9713-9721. doi:https://doi.org/10.1029/JC095iC06p09713 (1990).
Hennings, I., Matthews, J., & Metzner, M. Sun glitter radiance and radar cross-section modulations of the sea bed. 99(C8), 16303-16326. doi:https://doi.org/10.1029/93JC02777 (1994).
Hennings, I., Metzner, M., & Calkoen, C. J. Island connected sea bed signatures observed by multi-frequency synthetic aperture radar. International Journal of Remote Sensing, 19(10), 1933-1951. doi:10.1080/014311698215072 (1998).
Hernandez, W. J., & Armstrong, R. A. Deriving Bathymetry from Multispectral Remote Sensing Data. Journal of Marine Science and Engineering, 4(1). doi:10.3390/jmse4010008 (2016).
Hickman, G. D., & Hogg, J. E. Application of an airborne pulsed laser for near shore bathymetric measurements. Remote Sensing of Environment, 1(1), 47-58. doi:https://doi.org/10.1016/S0034-4257(69)90088-1 (1969).
Hsieh, S.-H., Liu, L.-W., Chung, W.-G., & Wang, Y.-M. J. W. Sensitivity analysis on the rising relation between short-term rainfall and groundwater table adjacent to an artificial recharge lake. 11(8), 1704. (2019).
Huang, M. Y. F., & Montgomery, D. R. Fluvial response to rapid episodic erosion by earthquake and typhoons, Tachia River, central Taiwan. Geomorphology, 175-176, 126-138. doi:https://doi.org/10.1016/j.geomorph.2012.07.004 (2012).
Jérôme, L., Gentile, V., Demarchi, L., Spitoni, M., Piégay, H., & Mróz, M. Bathymetric Mapping of Shallow Rivers with UAV Hyperspectral Data.(2016).
Janowski, L., Wroblewski, R., Rucinska, M., Kubowicz-Grajewska, A., & Tysiac, P. Automatic classification and mapping of the seabed using airborne LiDAR bathymetry. Engineering Geology, 301, 106615. doi:https://doi.org/10.1016/j.enggeo.2022.106615 (2022).
Kasvi, E., Laamanen, L., Lotsari, E., & Alho, P. Flow Patterns and Morphological Changes in a Sandy Meander Bend during a Flood—Spatially and Temporally Intensive ADCP Measurement Approach. Water, 9(2). doi:10.3390/w9020106 (2017).
Kasvi, E., Salmela, J., Lotsari, E., Kumpula, T., & Lane, S. N. Comparison of remote sensing based approaches for mapping bathymetry of shallow, clear water rivers. Geomorphology, 333, 180-197. doi:https://doi.org/10.1016/j.geomorph.2019.02.017 (2019).
Keim, R. F., Skaugset, A. E., & Bateman, D. S. Digital terrain modeling of small stream channels with a total-station theodolite. Advances in Water Resources, 23(1), 41-48. doi:https://doi.org/10.1016/S0309-1708(99)00007-X (1999).
Kim, J. S., Baek, D., Seo, I. W., & Shin, J. Retrieving shallow stream bathymetry from UAV-assisted RGB imagery using a geospatial regression method. Geomorphology, 341, 102-114. doi:https://doi.org/10.1016/j.geomorph.2019.05.016 (2019).
Kwon, S., Noh, H., Seo, I. W., & Park, Y. S. Effects of spectral variability due to sediment and bottom characteristics on remote sensing for suspended sediment in shallow rivers. Science of The Total Environment, 878, 163125. doi:https://doi.org/10.1016/j.scitotenv.2023.163125 (2023).
Lafon, V., Froidefond, J. M., Lahet, F., & Castaing, P. SPOT shallow water bathymetry of a moderately turbid tidal inlet based on field measurements. Remote Sensing of Environment, 81(1), 136-148. doi:https://doi.org/10.1016/S0034-4257(01)00340-6 (2002).
Lee, C. B., Traganos, D., & Reinartz, P. A Simple Cloud-Native Spectral Transformation Method to Disentangle Optically Shallow and Deep Waters in Sentinel-2 Images. Remote Sensing, 14(3). doi:10.3390/rs14030590 (2022).
Liu, L.-W., & Wang, Y.-M. Modelling reservoir turbidity using Landsat 8 satellite imagery by gene expression programming. Water, 11(7), 1479. (2019).
Liu, Y., Song, P., Peng, J., Fu, Q., Dou, C. J. H.-c. V., & Change. Recent increased frequency of drought events in Poyang Lake Basin, China: climate change or anthropogenic effects. 344, 99-104. (2011).
Loor, G. d. The observation of tidal patterns, currents, and bathymetry with SLAR imagery of the sea. IEEE Journal of Oceanic Engineering, 6(4), 124-129. doi:10.1109/JOE.1981.1145501 (1981).
Lu, S., Ouyang, N., Wu, B., Wei, Y., & Tesemma, Z. Lake water volume calculation with time series remote-sensing images. International Journal of Remote Sensing, 34(22), 7962-7973. doi:10.1080/01431161.2013.827814 (2013).
Lyzenga, D. R. Passive remote sensing techniques for mapping water depth and bottom features. Applied Optics, 17(3), 379-383. doi:10.1364/AO.17.000379 (1978).
Lyzenga, D. R. Remote sensing of bottom reflectance and water attenuation parameters in shallow water using aircraft and Landsat data. International Journal of Remote Sensing, 2(1), 71-82. doi:10.1080/01431168108948342 (1981).
Mandlburger, G., Kölle, M., Nübel, H., & Soergel, U. BathyNet: A Deep Neural Network for Water Depth Mapping from Multispectral Aerial Images. PFG – Journal of Photogrammetry, Remote Sensing and Geoinformation Science, 89(2), 71-89. doi:10.1007/s41064-021-00142-3 (2021).
Mandlburger, G., Pfennigbauer, M., Pfeifer, N. J. I. a. o. t. p., remote sensing, & sciences, s. i. Analyzing near water surface penetration in laser bathymetry—A case study at the River Pielach. 5, W2. (2013).
McFeeters, S. K. The use of the Normalized Difference Water Index (NDWI) in the delineation of open water features. International Journal of Remote Sensing, 17(7), 1425-1432. doi:10.1080/01431169608948714 (1996).
Mishra, D. R., Narumalani, S., Rundquist, D., & Lawson, M. Characterizing the vertical diffuse attenuation coefficient for downwelling irradiance in coastal waters: Implications for water penetration by high resolution satellite data. ISPRS Journal of Photogrammetry and Remote Sensing, 60(1), 48-64. doi:https://doi.org/10.1016/j.isprsjprs.2005.09.003 (2005).
Niroumand-Jadidi, M., Legleiter, C. J., & Bovolo, F. Bathymetry retrieval from CubeSat image sequences with short time lags. International Journal of Applied Earth Observation and Geoinformation, 112, 102958. doi:https://doi.org/10.1016/j.jag.2022.102958 (2022).
Pacheco, A., Horta, J., Loureiro, C., & Ferreira, Ó. Retrieval of nearshore bathymetry from Landsat 8 images: A tool for coastal monitoring in shallow waters. Remote Sensing of Environment, 159, 102-116. doi:https://doi.org/10.1016/j.rse.2014.12.004 (2015).
Parrot. Parrot Sequoia multispectral camera. Retrieved from https://www.parrot.com/assets/s3fs-public/2021-09/bd_sequoia_integration_manual_en_0.pdf. (2023).
Polcyn, F., Brown, W. L., & Sattinger, I. The measurement of water depth by remote sensing techniques. Retrieved from (1970).
Roberts, A. C. B. Shallow water bathymetry using integrated airborne multi-spectral remote sensing. International Journal of Remote Sensing, 20(3), 497-510. doi:10.1080/014311699213299 (1999).
Sagawa, T., Yamashita, Y., Okumura, T., & Yamanokuchi, T. Satellite Derived Bathymetry Using Machine Learning and Multi-Temporal Satellite Images. Remote Sensing, 11(10). doi:10.3390/rs11101155 (2019).
Sakimura, R., & Maruyama, K. Development of a New Generation Imaging Total Station System. 133(1), 14-22. doi:doi:10.1061/(ASCE)0733-9453(2007)133:1(14) (2007).
Samboko, H. T., Abas, I., Luxemburg, W. M. J., Savenije, H. H. G., Makurira, H., Banda, K., & Winsemius, H. C. Evaluation and improvement of remote sensing-based methods for river flow management. Physics and Chemistry of the Earth, Parts A/B/C, 117, 102839. doi:https://doi.org/10.1016/j.pce.2020.102839 (2020).
Sandidge, J. C., & Holyer, R. J. Coastal Bathymetry from Hyperspectral Observations of Water Radiance. Remote Sensing of Environment, 65(3), 341-352. doi:https://doi.org/10.1016/S0034-4257(98)00043-1 (1998).
Simpson, M. R. Discharge measurements using a broad-band acoustic Doppler current profiler: US Department of the Interior, US Geological Survey Reston, CA, USA.(2001).
Su, H., Liu, H., & Heyman, W. D. Automated Derivation of Bathymetric Information from Multi-Spectral Satellite Imagery Using a Non-Linear Inversion Model. Marine Geodesy, 31(4), 281-298. doi:10.1080/01490410802466652 (2008).
Tamminga, A., Hugenholtz, C., Eaton, B., & Lapointe, M. Hyperspatial Remote Sensing of Channel Reach Morphology and Hydraulic Fish Habitat Using an Unmanned Aerial Vehicle (UAV): A First Assessment in the Context of River Research and Management. River Research and Applications, 31(3), 379-391. doi:https://doi.org/10.1002/rra.2743 (2015).
Tomasi, C., & Kanade, T. J. I. J. C. V. Detection and tracking of point. 9, 137-154. (1991).
Tonina, D., McKean, J. A., Benjankar, R. M., Wright, C. W., Goode, J. R., Chen, Q., . . . Edmondson, M. R. Mapping river bathymetries: Evaluating topobathymetric LiDAR survey. Earth Surface Processes and Landforms, 44(2), 507-520. doi:https://doi.org/10.1002/esp.4513 (2019).
Traore, S., Luo, Y., & Fipps, G. Gene-Expression Programming for Short-Term Forecasting of Daily Reference Evapotranspiration Using Public Weather Forecast Information. Water Resources Management, 31(15), 4891-4908. doi:10.1007/s11269-017-1784-5 (2017).
Van-An, N., Hsuan, R., Chih-Yuan, H., & Kuo-Hsin, T. Bathymetry derivation in shallow water of the South China Sea with ICESat-2 and Sentinel-2 data. Journal of Applied Remote Sensing, 15(4), 1-17. doi:10.1117/1.JRS.15.044513 (2021).
Vinod Kumar, K., Palit, A., & Bhan, S. J. I. J. o. R. S. Bathymetric mapping in Rupnarayan-Hooghly river confluence using Indian remote sensing satellite data. 18(11), 2269-2270. (1997).
Wu, C.-H., Chen, S.-C., & Chou, H.-T. Geomorphologic characteristics of catastrophic landslides during typhoon Morakot in the Kaoping Watershed, Taiwan. Engineering Geology, 123(1), 13-21. doi:https://doi.org/10.1016/j.enggeo.2011.04.018 (2011).
于瑞宏, 刘廷玺, 李畅游, & 许有鹏. 干旱区草型湖泊悬浮固体浓度及水深的遥感与分析. 水利学报, 36(7), 853-862. (2005).
王艳姣, & 张鹰. 基于 BP 人工神经网络的水体遥感测深方法研究. 海洋工程, 23(4), 33-38. (2005).
许殿元. 黄河口海域水深分布的遥感研究. 遥感技术与应用, 7(3), 17-23. (1992).
沈芳, & 匡定波. 青海湖最近 25 年变化的遥感调查与研究. 湖泊科学, 15(4), 289-296. (2003).
武彦斌, 彭苏萍, 黄明, & 邹冠贵. 基于神经网络的沉陷区水深遥感研究. 煤田地质与勘探(2), 41-44. (2007).
陈文革, 黄铁侠, & 卢益民. 机载海洋激光雷达发展综述. 激光技术, 22(3), 147-152. (1998).
陈坚, & 金翔龙. 机载激光测深技术进展及应用. 海洋通报, 21(6), 75-82. (2002).
陈烽. 近海机载激光海洋测深技术. 应用光学, 20(2), 18-23. (1999).
赵建虎, 欧阳永忠, & 王爱学. 海底地形测量技术现状及发展趋势. (2017).
赵英时. 遥感应用分析原理与方法: 科学出版社.(2013).
高屏溪流域管理委員會. 108 年高屏溪流域管理工作執行年報. (2020).
張斐章, & 張麗秋. 類神經網路導論：原理與應用: 滄海.(2010).
黄韦艮, 傅斌, 周长宝, 杨劲松, 史爱琴, & 厉冬玲. 星载 SAR 水下地形和水深遥感的最佳雷达系统参数模拟. 遥感学报, 4(3), 172-177. (2000).
經濟部水利署. 經濟部水利署施工規範第01726章水深測量. Retrieved from https://wralaw.wra.gov.tw/Law_Content.aspx?n=9334&s=12033&sms=9191. (2022).
蔡耀誼. 應用基因表達規劃法與遺傳演算法調控箱型理論於股票市場之研究. (碩士). 輔仁大學, 新北市. Retrieved from https://hdl.handle.net/11296/3n95sr (2016).
鄧連才. 應用地理資訊系統與類神經網路於模擬沙灘海岸線之關係-以墾丁國家公園為例. (碩士). 國立屏東科技大學, 屏東縣. Retrieved from https://hdl.handle.net/11296/8zs497 (2012).