Machine Learning-Based High-Resolution Estimation of Global Catch Distribution of Taiwan’s Distant Water Fisheries

@inproceedings{Huang2023MachineLearningBased,
  author = {Huang, Er-Qun and Hsu, William W.Y. and Weng, Ming-Wei},
  booktitle = {2023 IEEE 5th Eurasia Conference on IOT, Communication and Engineering (ECICE)},
  title = {Machine Learning-Based High-Resolution Estimation of Global Catch Distribution of Taiwan’s Distant Water Fisheries},
  year = {2023},
  volume = {},
  number = {},
  pages = {620-622},
  abstract = {Taiwan’s overseas fishing fleet is well-known for its size and operation efficiency. Due to technological advancement, Taiwan has an excellent coastal and offshore mapping system of precision of 0.01 and 0.001° in locating fishing grounds and plotting catch distribution. However, for overseas fisheries, traditional methods for mapping catch distribution use only the trajectory of the fishing vessel with eLogbooks to distribute the catch on 1.0° using the geographic information system (GIS) [1]. Thus, we developed an approach to map catch distributions of Taiwan’s overseas fishing fleet using artificial intelligence (AI) and high-resolution geographic information system (GIS) technology in 0.5 and 0.25°. We integrated vessel monitoring system (VMS) data and eLogbooks records to recognize fishing activities and improve the accuracy of fishing locations. We combined the VMS data and eLogbook catch records of Taiwan fishing vessels to recognize the species of catch of each fishing vessel. After obtaining the species for each fishing vessel, we used machine learning to identify fishing and non-fishing areas to assign weighted catches using a machine learning algorithm. The result was analyzed globally on a high-resolution scale of 0.25 °. The findings showed that the system mapped the catch volumes of different fish species and differentiated between fishing and non-fishing areas while the fishing vessel sailed. This effectively determined catch distributions in areas where vessels were not fishing, providing more accurate fishing areas and enhancing resource assessment. In addition, a distinctive feature to visualize the distinct catch and species in a high resolution was provided.},
  keywords = {Machine learning algorithms;Sea measurements;Machine learning;Fish;Trajectory;Planning;Water resources;distant water fishing;catch distribution;fisheries resource management;vessel monitoring system;eLogbook},
  doi = {10.1109/ECICE59523.2023.10382986},
  issn = {},
  month = {Oct},
}