测绘科学技术学报2011，Vol.28Issue(1)：70-74,78,6.DOI:10.3969/j.issn.1673-6338.2011.01.017
利用水深不确定度探测测深异常值的方法
The Approach on Detecting Outliers of Multi-Beam Data by Uncertainty
摘要
Abstract
The seafloor surface could be constructed effectively by LS-SVM depend on the train samples chosen reasonably. In the process of seafloor surface conformation, a new method of optimize samples by uncertainty was presented. Some practical multi-beam data was chosen to verify the correctness and rationality of the new method. The results between trend surface filter and the new method showed that the effective train samples could be chosen and the influence of the sample-outliers could be restrained. The reasonable seafloor surface could be constructed by LS-SVM arithmetic, and then the outliers of multi-beam data could be eliminated effectively.关键词
最小二秉支持向量机(LS-SVM)/趋势面滤波/不确定度/测深异常值/训练样本Key words
LS-SVM/ trend surface filter/ uncertainty/ sounding outliers/ train samples分类
天文与地球科学引用本文复制引用
黄贤源,翟国君,黄谟涛,隋立芬,柴洪洲,陆秀平..利用水深不确定度探测测深异常值的方法[J].测绘科学技术学报,2011,28(1):70-74,78,6.基金项目
国家863计划资助项目(2007AA12Z326) （2007AA12Z326）
国家自然科学基金资助项目(40974010). （40974010）
