The establishment of automatic detection procedures for far and near outliers in multibeam echo sounding dataset

Abstract

Multibeam echosounder (MBES) has tremendously improved the rate of data collection in terms of time spend and data density. MBES collects full data coverage of seabed within significantly short period of time. However more time is required for data cleaning process in post-processing mode (Cronin et al., 2003). The data cleaning process is an important event as the collected data comprised of not only real seabed profiles but associated with erroneous data called outliers. The integrity of acquired data could be validated if a ground truth validation could be performed. However a systematic checking on the grounds of multibeam data is impossible (Mori, 2003). This technical paper discusses on the development of automatic detection of MBES outliers which is divided into two categories namely far outliers and near outliers. From various programs developed using Microsoft Visual Basic, the study tries to investigate and establish the least procedures to detect MBES outliers.