Automatic Identification of Fish Species and Their Farmed or Wild Origin by Computer Vision and Deep Learning
摘要
This work is part of the GLORiA research project and focuses on the use of deep learning and computer vision in aquaculture. Specifically, this work deals with the automatic identification of escaped fish. The aim is to use images of the fish, previously acquired in the laboratory with pre-established patterns, to obtain information about the species it belongs to, whether it is farmed, escaped or wild. We will use a pre-trained network, ResNet50, in combination with transfer learning techniques for this purpose. In addition, we will also develop a prototype application. This application will integrate the trained model to provide a practical tool for automated identification and classification of fish based on uploaded images.