岸外沙洲潮间带地形的增强型遥感构建方法

水边线法是潮间带地形遥感反演的重要方法。针对常规水边线方法在应用于地形多变的岸外沙洲潮间带数字高程模型（DEM）构建中出现的水边线交叉和缺乏表现潮沟微地形特征的问题，本文提出了一种增强型地形遥感构建方法:首先通过潮位排序，筛选出具有正确高程变化趋势、空间分离的水边线；然后对筛选过程中水边线位置相近的影像组按像元计算改进的归一化差值水体指数并进行均值序列合成，提取出合成海陆边界线，二者共同构建出初始的潮间带DEM；进一步利用低潮期的潮沟边界线和中线生成潮沟DEM，通过潮沟镶嵌处理，最终获得可表达滩面微地形起伏的潮滩DEM。该方法在江苏岸外辐射沙洲中部核心区域的模拟应用表明，对比4条验证剖面的平均结果，高程平均绝对误差为0.43 m，均方根误差为0.54 m，相关系数（r）为0.75，模拟高程与验证高程在剖面起伏形态变化方面具有良好的一致性，同时模拟DEM的空间破碎度小，能够反映更多的细节地形特征。该方法可为利用多源遥感数据构建考虑微地形变化的高精度潮间带地形DEM提供新的思路。

The enhanced construction method for intertidal terrain of offshore sandbanks by remote sensing

The waterline method is an important method for remote sensing inversion of intertidal terrain. Aiming at the problems in the digital elevation model (DEM) construction with variable topography, which the waterlines intersect and lack of representation of micro-topography for the tidal creek in the conventional waterline method, an enhanced terrain remote sensing construction method was proposed in this paper. First, the waterlines with the correct trend of elevation change and spatial separation were selected through tidal level sorting. Second, the images group with similar waterlines position in the screening process were combined according to the mean of a sequence of the modified normalized difference water index values in each pixel, and the boundary line of land-sea was obtained. These two methods worked together to construct the initial intertidal DEM. Then, the tidal creek DEM was generated by using the boundary and centerline of the tidal creek in the low tide period. Through the mosaic processing, the tidal flat DEM that can express the undulating micro-topography of the beach surface was constructed. Comparing with the measured terrain, the results in the central core area of the radial sand ridges off the coast of Jiangsu showed that the mean absolute error of the 4 measured sections was 0.43 m, the root mean square error was 0.54 m, and the average correlation (r) was 0.75. The simulation results and the measured results were in good agreement with the changes in beach undulations. Furthermore, the spatial fragmentation of the simulated DEM was small and can reflect more detailed terrain features. This method can provide a new idea for constructing high-precision DEM of intertidal zone considering micro terrain changes by using multi-source remote sensing data.