Detection and Classification of Corals from Under-Sea Images using Deep Learning

Tapaswini Nayak; Bhagyashree Patra

doi:10.36948/ijfmr.2025.v07i05.59219

Detection and Classification of Corals from Under-Sea Images using Deep Learning

Author(s)	Dr. Tapaswini Nayak, Ms. Bhagyashree Patra
Country	India
Abstract	Coral reefs are considered, an essential part of the marine ecosystem, as they support a large part of marine life. Deep sea exploration and imaging have provided a great opportunity to look into the vast and complex marine ecosystem. Manual annotation is a tedious and time- consuming task for marine experts and takes a lot of time for them to manually annotate a single image. Therefore, automatic detection of corals is important as it helps the experts to keep track of vulnerable coral species in conjunction with other marine life, to detect and annotate them. However, this is a challenging task, as most of the common coral species look familiar to each other even if they belong to a different category. Additionally, in a single colony, varieties of coral co-exhibit and coral reefs serve as the home of a variety of aquatic species. Since, deep learning procedures of detection and classification are based on photographic images with features like image quality, distance from the image, angle of the object and underwater lighting conditions, this task is challenging. A large dataset of 2500 images has been taken and is classified into 10 different coral species in the present automated deep-learning coral detection scheme, for detection and classification accuracy. In the current work, the fundamental analysisof these images is done, using techniques of Deep learning (DL) and Convolutional Neural Networks (CNN). To enhance the easiness, efficiency and speed, a known database of corals called structure RSMAS (consisting only 600 images) is used to develop/design a prototype. To enhance the capability of our system, public image data available online at inaturalist.org, with 1000 images has been used, giving an accuracy of 94%. Currently, an existing DL model called as MobileNet (for classification) with an SSD detection head (for detection), is being used. Further, detection and classification could be increased by data cleaning and using high- resolution images and design a customized DL model Architecture.
Keywords	Coral Detection, Structural Coral images, Deep Learning, Transfer Learning, Convolutional Neural Network, Structured RSMAS, MobileNet, SSD.
Field	Computer
Published In	Volume 7, Issue 5, September-October 2025
Published On	2025-10-31
DOI	https://doi.org/10.36948/ijfmr.2025.v07i05.59219

View / Download PDF File

E-ISSN 2582-2160

doi

CrossRef DOI is assigned to each research paper published in our journal.

IJFMR DOI prefix is
10.36948/ijfmr

Downloads

Research Paper Format Copyright Permission Form and Undertaking Form Cover Page Vol 8 Isu 2 Cover Page Vol 8 Isu 1 Cover Page Vol 7 Isu 6

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Detection and Classification of Corals from Under-Sea Images using Deep Learning

Share this