在验证趋势面滤波是最小二乘支持向量机算法(LS-SVM)取特定参数解的基础上,利用LS-SVM所构造的海底趋势面对测深异常值进行剔除。为了克服LS-SVM解非稀疏性的缺点,同时抑制偏差较大的训练样本对海底趋势面构造的影响,提出一种基于局部样本中心距离的训练样本优化方法。为了检验该算法的有效性,选取实测的多波束测深数据进行验证,结果表明在训练样本优化的基础上,通过调整LS-SVM的参数可以得到更为合理的海底趋势面,测深异常值地剔除也更为有效。 更多还原
【Abstract】 As validating the trend filter is the special result to the LS-SVM arithmetic,the sounding outliers are eliminated by the seafloor surface which constructed by LS-SVM.In order to solve the sparseness of LS-SVM results and restrain the influence of the sample-outliers,a new method of optimized samples by part samples center distance is presented.Some practical multi-beam data is chosen to verify the correctness and rationality of the new method.The example shows that on the ground of the optimize... 更多还原