Support vector machine classification method for predicting Jakarta bay bottom sediment type using multibeam echosounder data
2020
Manik, Henry Munandar | Solikin, Steven | Sri Pujiyati | Susilohadi
The need for accurate seafloor maps is increasing along with the increase in marine activities, such as dredging, construction of buildings on the coast and offshore, and navigation of ships to prevent shipwrecks while sailing. The hydroacoustic technology used in this study is the multibeam echosounder system (MBES), which is the most advance acoustic instrument today. MBES can sweep very large areas in a short time, so that the survey costs can be reduced. The aim of this research was firstly to classify the seabed sediment in G-Island, Jakarta Bay using supervised classification technique. Secondly, to analyze the acoustic characteristic of the seabed sediment and compare it with the physical characteristic of the sediment.This research was conducted on October 31st to November 5th 2016 in the waters of G-Island, Jakarta Bay. In this study, supervised classification techniques were applied. The supervised classification techniques used in this research was Support Vector Machine (SVM). SVM produces classifications with 5 main classes, namely clay, fine silt, medium silt, coarse silt and fine sand. The overall accuracy value of the SVM method was 80.25% with the Kappa coefficient value of 0.2031 which is categorized into the fair class in its classification.
Afficher plus [+] Moins [-]Mots clés AGROVOC
Informations bibliographiques
Cette notice bibliographique a été fournie par Universiti Putra Malaysia
Découvrez la collection de ce fournisseur de données dans AGRIS