Paper Title
Using Gene-Expression Programming on Bathymetry Modeling – A Case Study of Taiwan Mountainous River

Abstract
Shallow bathymetry investigation is essential for river engineering, water resources, and coastal area management. Since the investment of advanced remote sensing technologies, such as satellites and drones, the investigator can conduct the survey work in a large area within a short period. These remote sensing technologies require a bathymetry retrieval model to convert the imagery information to bathymetry. However, due to the river channel in Taiwan being shallow and narrow, the application of existing models in Taiwan has essentially difficult. Therefore, this study aimed to establish a shallow bathymetry retrieval model fitted with Taiwan’s river channel condition. Therefore, this study aimed to establish a model suitable for mountainous river areas utilizing an unmanned aerial vehicle (UAV) equipped with a multispectral camera and machine learning-based gene-expression programming (GEP) algorithm. Themultispectral data were retrieved from themosaicedimage contents an average of 207.9 million 3D densified points,and combined with a total of 171 water depth measurements for bathymetry modeling. The results show that the coefficient of determination (R2) of GEP is 0.801, the mean absolute error (MAE) is 0.154 m, and root mean square error (RMSE) is 0.195 m. The model performance of GEP model has increased by 16.3% in MAE, compared to conventional simple linear regression (REG) algorithm, and has a lower bathymetry retrieval error both in shallow (<0.4 m) and deep waters (>0.8 m). The GEP bathymetry retrieval model has a considerable degree of accuracy and could be applied to shallow rivers or near-shore areas under similar conditions of this study. Keywords - River Survey; Multispectral Camera; Unmanned Aerial Vehicle (UAV)