Title: The intelligent object detection framework for detecting fish from underwater images

Authors: Kalyani Peddina; Ajay Kumar Mandava

Addresses: Department of Electrical Electronics and Communication Engineering, GITAM Bengaluru Campus, NH 207, Nagadenehalli Doddaballapur, Taluk, Bengaluru, Karnataka 561203, India ' Department of Electrical Electronics and Communication Engineering, GITAM Bengaluru Campus, NH 207, Nagadenehalli Doddaballapur, Taluk, Bengaluru, Karnataka 561203, India

Abstract: Marine applications heavily rely on underwater object detection, yet challenges like complex backgrounds and image quality issues impede deep learning-based detectors. Monitoring feed pellet utilisation in aquaculture is vital for efficient resource management. This study introduces a novel framework, DYNFS, merging underwater object detection and image reconstruction using YOLO-V5. Initially, we curate an underwater image dataset, refining it to remove noise, and then employ DYNFS for classification. Our approach achieves a 98.93% accuracy rate in identifying submerged feed pellets, crucial for aquaculture efficiency. However, locating pellets remains challenging due to poor image quality and small object sizes. The enhanced YOLO-V5 networks show promise in real-world aquaculture scenarios. This framework enhances underwater object detection, offering potential benefits for marine applications and aquaculture management.

Keywords: detection; image data; convolutional neural network confusion matrix; YOLO network; recurrent neural network.

DOI: 10.1504/IJCNDS.2025.142996

International Journal of Communication Networks and Distributed Systems, 2025 Vol.31 No.1, pp.63 - 88

Received: 12 Sep 2023
Accepted: 17 Jan 2024
Published online: 02 Dec 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article