【摘要】 基于Landsat8 OLI传感器数据,用BP人工神经网络模拟水深法和底部反照率独立水深测量算法(bottom albedo-independent bathymetry algorithm,简称B算法)来反演黄河口水深,并与实测水深数据进行比较;然后,基于两种水深反演方法的结果进行了对比分析和适用性评价.发现,用BP神经网络模拟方法提取的水深在近岸水深小于15m的区域和水深大于15m的区域,反演结果与实测水深误差较大,变化趋势不一致,实测值与模拟水深值相关系数低;以B算法提取的水深与实测水深误差较小,在趋势上一致,相关系数为0.899;并基于该算法的表现,通过水深遥感制图进一步在实际应用中验证了该水深反演方法.结果表明,底部反照率独立测深算法精度高、效果好,比较适用于黄河口水深探测.更多还原

【Abstract】 Back propagation Artificial Neural Network(BP-ANN)method and Bottom Albedo-independent Bathymetry Algorithm(B algorithm)were used to derive the water depth in Yellow River Estuary based on the Landsat 8OLI sensor’s data in this paper.Besides,the derived data and in-situ measurement data were compared and analyzed,from which we find that:there have error-large and change trend-inconsistencies problems during the BP-ANN derive and in-situ measurement water depth in the areas the water depth within and above 10m;while B algorithm has a good performance on the error and change trends aspects,especially in the area which the water depth within and above 10 meters.In addition,the correlation coefficients between B algorithm derive and in-situ measurement data was 0.899.Mapping the water depth is a further proof of B algorithm based on the good performance of which.The results show that B algorithm which have high accuracy and better effect is suitable approach for the water depth derive in the Yellow River Estuary.更多还原