A fast algorithm for detecting deep sea transshipment activities using discrete GPS signals

碩士 === 國立臺灣海洋大學 === 資訊工程學系 === 106 === Illegal, unreported and unregulated (IUU) fishing activity haven been attracting attention in recent years. Moreover, some species of fish are endangered and can be extinct at any moment in recent future. To battle IUU, the Food and Agriculture Organization of...

Full description

Bibliographic Details
Main Authors: Lu, Min-Hsuan, 呂旻軒
Other Authors: Hsu, William W.Y.
Format: Others
Language:zh-TW
Published: 2018
Online Access:http://ndltd.ncl.edu.tw/handle/hjc97q
Description
Summary:碩士 === 國立臺灣海洋大學 === 資訊工程學系 === 106 === Illegal, unreported and unregulated (IUU) fishing activity haven been attracting attention in recent years. Moreover, some species of fish are endangered and can be extinct at any moment in recent future. To battle IUU, the Food and Agriculture Organization of the United Nations approved the plan of the International Plan Of Action To Prevent, Deter And Eliminate Illegal, Unreported And Unregulated Fishing (IPOA-IUU). To support eliminating IUU, regional fisheries management organizations (RFMO) and countries strive to establish a variety of protection measures which can make fisheries resources sustainable. Transshipment activities are publicly known vulnerable behavior which can cause IUU activities. Taiwan government enacted laws and regulations to ban IUU fishing activities by requiring deep-sea operating vessels to report transshipments. However, fishing vessels can still conduct illegal and unreported transshipments by not reporting. We know that under regulations, fishing vessels must report Vessel Monitoring System (VMS) data on a regular basis. The objective of this research wants to find which vessels are potentially conducting transshipment activities based on their reported VMS data. Since there are many vessels and VMS data to be computed, we have developed a fast algorithm to identify which pairs of vessels have potentially contacted within an effective time frame for law enforcement. In addition, we have also developed a module for simulating the trajectory of the vessel pair for further human analysis. By comparing the results of our analysis and the records held by the Fisheries Agency of Taiwan, our algorithm has an accuracy over 90%. However, our algorithm has not considered vessels which are contacting for supplying and fueling activities. Future works will take this into account.