Poaching Surveillance System using Class B AIS Units

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K04576
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 25020:Safety engineering-related
• Research Institution: National Institute of Technology(KOSEN), Oshima College
• Principal Investigator: Okamura Kenshiro 大島商船高等専門学校, 情報工学科, 嘱託教授 (60194388)
• Co-Investigator(Kenkyū-buntansha): 松村 遼 周南公立大学, 福祉情報学部, 准教授 (20734768)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 密漁監視 / AIS / 不審船 / 漁業許可船 / 画像処理

Outline of Final Research Achievements

In recent years, illegal fishing in coastal area has become a significant social issue. Usually, illegal fishing involves the use of conventional fishing vessels. So, it is difficult to detect automatically illegal vessels using image processing based on vessel appearance and navigation patterns. Therefore, this study estimates the positions of vessels captured by surveillance cameras and compares these positions with those obtained from AIS (Automatic Identification System) signals to identify licensed fishing vessels. Subsequently, a method detects suspicious vessels by excluding identified licensed vessels. To implement this method, we developed a system using fixed-point cameras to estimate vessel positions and sizes in real-world scenarios. Additionally, we established a mechanism to facilitate the collection of camera calibration data and training images for AI-based image recognition via AIS signals.

Free Research Field

画像処理

Academic Significance and Societal Importance of the Research Achievements

沿岸域に生息するアワビ，ナマコ等は容易に採取できるだけでなく莫大な利益をもたらすことから，組織的かつ広域的な密漁が横行し，大きな社会問題となっている．しかしながら一般の漁船が密漁に用いられるためその外観や運航軌跡を用いた自動監視システムの構築は非常に難しい．
そこで，本研究においては，密漁船を画像処理により直接見つけるのではなく，監視領域において検出した船舶の中から進入許可船を見つけ，これらを取り除いた船舶を密漁船と考えるという発想転換を行った．進入許可船の判断にはAISを用いるため，単に密漁対策のためだけではなく海上交通の安全にも寄与するシステムとなっている．