Deep WaveNet-based YOLO V5 for Underwater Object Detection

Adithya Balaji, S. Yogesh, C. K. Kalyaan, R. Narayanamoorthi, Gerard Dooly, Samiappan Dhanalakshmi

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Detecting objects under the water is always a difficult task. Due to the unique characteristics of underwater settings, identifying underwater objects can be difficult. When light penetrates deep within the water, the aqueous medium exhibits diffraction and scattering. This leads to murky, obscured films and photos, which make interpretation more difficult. Underwater object detection has become one of the most important objectives in deep-sea exploration. To overcome the above issues, we propose a methodology based on the YOLO model in which the images will be passed through for detection. The images are trained using the CNN-based deep wave net algorithm which enhances the faded or distorted image further making the YOLO algorithm infer better results. The YOLO models are compared with other detection models and different activation layers are also compared.

Original languageEnglish
Title of host publicationOCEANS 2023 - Limerick, OCEANS Limerick 2023
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798350332261
DOIs
Publication statusPublished - 2023
Event2023 OCEANS Limerick, OCEANS Limerick 2023 - Limerick, Ireland
Duration: 5 Jun 20238 Jun 2023

Publication series

NameOCEANS 2023 - Limerick, OCEANS Limerick 2023

Conference

Conference2023 OCEANS Limerick, OCEANS Limerick 2023
Country/TerritoryIreland
CityLimerick
Period5/06/238/06/23

Keywords

  • Deep Learning
  • Neural Networks
  • Object Detection

Fingerprint

Dive into the research topics of 'Deep WaveNet-based YOLO V5 for Underwater Object Detection'. Together they form a unique fingerprint.

Cite this