Research Article
Developing an Efficient Underwater Object Detection System using AI
@INPROCEEDINGS{10.4108/eai.28-4-2025.2357814, author={Alla Nithin Reddy and Mendu Sravanthi and Shaik Dariya Hussain and Aluri Aditya Choudary and Brij Kishor Tiwari}, title={Developing an Efficient Underwater Object Detection System using AI}, proceedings={Proceedings of the 4th International Conference on Information Technology, Civil Innovation, Science, and Management, ICITSM 2025, 28-29 April 2025, Tiruchengode, Tamil Nadu, India, Part I}, publisher={EAI}, proceedings_a={ICITSM PART I}, year={2025}, month={10}, keywords={underwater object detection deep learning yolo image enhancement aquarium dataset real-time detection}, doi={10.4108/eai.28-4-2025.2357814} }- Alla Nithin Reddy
Mendu Sravanthi
Shaik Dariya Hussain
Aluri Aditya Choudary
Brij Kishor Tiwari
Year: 2025
Developing an Efficient Underwater Object Detection System using AI
ICITSM PART I
EAI
DOI: 10.4108/eai.28-4-2025.2357814
Abstract
Underwater object detection is essential for ocean investigation, ecological monitoring, and security systems. However, challenges such as light absorption, scattering, and limited visibility significantly undermines detection accuracy. In this work, we present a robust deep learning-based approach that leverages advanced YOLO variants such as YOLOv9, YOLOv10, YOLOv11, and YOLOv12 for precise detection of underwater objects identification. For further improvement of detection performance, we integrate image enhancement techniques that mitigate underwater deformations and enhance feature extraction. The models are developed and trained and tested with the aquarium dataset, which provides diverse and realistic underwater images. Experimental results show that our enhanced pipeline significantly increases detection accuracy. This work contributes to the development of reliable underwater object detection systems in challenging aquatic environments.
