Identifying species of common sea fish harvested by longliner using deep convolutional neural networks
Journal
2019 ASABE Annual International Meeting
Date Issued
2019-01-01
Author(s)
Lu, Yi Chin
Abstract
© 2019 ASABE Annual International Meeting. All rights reserved. Fish catch statistics reported by vessels are essential information for the management of marine resource. The statistics, including species, body lengths, harvest time, and harvest location, were conventionally recorded by observers or fishermen. Manual recording is time consuming and can be subjective; thus, the accuracy of the statistics is questionable. The decks of fishing vessels are usually full of miscellaneous items, making automatic reporting of the statistics challenging. In recent years, convolutional neural networks (CNNs) have become increasingly popular and been applied to solving complex machine vision tasks. This study proposed to automatically identify 11 species or types of fish harvested by longliners using deep CNNs. The species included albacore (Thunnus alalunga), bigeye tuna (T. obesus), yellowfin tuna (T. albacares), southern bluefin tuna (T. maccoyii), blue marlin (Makaira nigricans), Indo-Pacific sailfish (Istiophorus platypterus), swordfish (Xiphias gladius), and dolphin fish (Coryphaena hippurus). Four deep CNNs, VGG-16, ResNet-50, DenseNet-201, MobileNetV2, were modified to identify the species of the fish. The CNNs outperformed conventional machine learning approaches and reached an accuracy of as high as 95.82% in species identification.
Subjects
Convolutional neural network | Deep learning | Fine-grained classification | Fish species identification | Model visualization
SDGs
Other Subjects
Convolution; Deep learning; Fish; Fishing vessels; Harvesting; Information management; Marine biology; Neural networks; Conventional machines; Convolutional neural network; Coryphaena hippurus; Fine grained; Fish species; Model visualization; Southern bluefin tuna; Species identification; Deep neural networks
Type
conference paper