Data analytics of the integrated fisheries informatic database using information and communication technologies

View Statistics Email Alert RSS Feed

Information

Details

Project title

Code/計畫編號

109農科-9.1.5-子-F1

Translated Name/計畫中文名

資通訊科技應用於漁業資料整合系統之研究

Funding Organization/主管機關

Council of Agriculture,Executive Yuan

Co-Investigator(s)/共同執行人

陳科仰(計畫主持人)

Department/Unit

Planning Division,Fisheries Agency,COA,Executive Yuan

Website

https://www.grb.gov.tw/search/planDetail?id=13674188

Year

2020

Start date/計畫起

01-01-2020

Expected Completion/計畫迄

31-12-2020

Co-Investigator(s)

William Wei-Yuan Hsu

Bugetid/研究經費

5300千元

ResearchField/研究領域

漁業

因應歐盟的黃牌警告，行政院農委會漁業署於2015年開始，補助開發新世代海洋資訊系統。截至目前為止，這套系統除了整合國內眾多漁業資訊外，也同時納入了AIS、以及多方資料來源的氣象、海象、海洋保護區、危險海域等資訊，並可即時對監控之船隻進行預警。隨著政府想了解與管理海上漁船及漁工之工作狀況，本計畫將利用電子漁獲日誌(eLogbook)與漁船航程回報器(VMS)之訊號以機器學習之方式進行解析。計畫中將設計出的智慧型演算法會先判斷eLogbook回傳資料的正確性並依照常規且合理的修正人為錯誤，結合不同來源之eLogbook資料，如衛星狀態不穩時手動回報以及圍網船所使用之iFIMS系統，進而與VMS之資料進行最佳之匹配。同時間系統將依照漁船資訊以及各層面之資料學習其形態，以eLogbook回報之細部資訊配合，智慧型演算法將估算出漁工可能之作業時數，同時計算捕獲漁獲之分佈。對於管理上，將強化資料視覺化界面，使得資料呈現更為清晰明朗。且將各模組與戰情中心連結起來，呈現最即時之摘要資訊，提供前線人員決策之使用。 The Fisheries Agency of Taiwan supported the development of the next generation oceanic system since 2015 to assist in the lifting of the yellow card warning posted by the European Commission. Over the course 4 years, this system has grown to include not only information related to fisheries but also the automatic identification system (AIS), weather data, ocean data, marine protected area information, and high-risk areas on the seas. Early warning capabilities of the system is possible due to the vast amount of data collected. Since the government has always been concerned about the status of fishing vessels and crews on seas, the main objective project focuses on using the eLogbook system and the VMS system with the help of machine learning techniques to understand the condition. Our proposed subsystem will intelligently correct wrongfully reported eLogbooks according to common sense and reasoning algorithms, and combine different sources of eLogbooks, i.e., manually reported modes due to satellite instability and the iFIMS system used by purse seiners. The intelligent algorithm will try to learn the behavior of fishing vessels from different information aspects combined with the detail parameters reported in the eLogbooks to compute the possible workload of fishers and the distribution of catches. Finally, in the managerial aspect, we will upgrade the linkage between each module and the combat intelligence center (CIC) to provide frontline managers with the most up-to-date information to assist their decision making.

Keyword(s)

海洋資訊科學
雲端架構
人工智慧預測
漁業管理
電子化行政
電子漁獲日誌
巨量資料
Oceanic information science
Cloud architecture
rtificial intelligent predicitions
Fisheries management
Electronic administration
Electronic logbooks
Big data

DSpace CRIS

Data analytics of the integrated fisheries informatic database using information and communication technologies

Details

Description