Citation: Li, Y.; Bai, X.; Xia, C. An
Improved YOLOV5 Based on Triplet
Attention and Prediction Head
Optimization for Marine Organism
Detection on Underwater Mobile
Platforms. J. Mar. Sci. Eng. 2022, 10,
1230. https://doi.org/10.3390/
jmse10091230
Academic Editors: Simone Marini,
Jacopo Aguzzi, Giacomo Picardi,
Damianos Chatzievangelou,
Sascha Flögel, Sergio Stefanni,
Peter Weiss and Daniel Mihai Toma
Received: 27 July 2022
Accepted: 30 August 2022
Published: 2 September 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Journal of
Marine Science
and Engineering
Article
An Improved YOLOV5 Based on Triplet Attention and
Prediction Head Optimization for Marine Organism Detection
on Underwater Mobile Platforms
Yan Li
1,2,
* , Xinying Bai
1,2,3
and Chunlei Xia
4,
*
1
State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences,
Shenyang 110016, China
2
Institutes of Robotics and Intelligent Manufacturing, Chinese Academy of Sciences, Shenyang 110169, China
3
College of Information, Liaoning University, Shenyang 110136, China
4
Yantai Institute of Coastal Zone Research, Chinese Academy of Sciences, Yantai 264003, China
* Correspondence: liyan1@sia.cn (Y.L.); clxia@yic.ac.cn (C.X.)
Abstract:
Machine vision-based automatic detection of marine organisms is a fundamental task for
the effective analysis of production and habitat changes in marine ranches. However, challenges of
underwater imaging, such as blurring, image degradation, scale variation of marine organisms, and
background complexity, have limited the performance of image recognition. To overcome these issues,
underwater object detection is implemented by an improved YOLOV5 with an attention mechanism
and multiple-scale detection strategies for detecting four types of common marine organisms in the
natural scene. An image enhancement module is employed to improve the image quality and extend
the observation range. Subsequently, a triplet attention mechanism is introduced to the YOLOV5
model to improve the feature extraction ability. Moreover, the structure of the prediction head of
YOLOV5 is optimized to capture small-sized objects. Ablation studies are conducted to analyze
and validate the effective performance of each module. Moreover, performance evaluation results
demonstrate that our proposed marine organism detection model is superior to the state-of-the-art
models in both accuracy and speed. Furthermore, the proposed model is deployed on an embedded
device and its processing time is less than 1 s. These results show that the proposed model has the
potential for real-time observation by mobile platforms or undersea equipment.
Keywords:
marine organism; target identification; deep learning; attention mechanism; model
optimization
1. Introduction
Underwater organism observation is an important topic in the field of underwater
object detection, which can offer an effective means to evaluate the abundance of marine or-
ganisms and sensitively predict environmental changes. For instance, it can autonomously
and intelligently identify and analyze the number of sea cucumbers, scallops, and other
seafood, as well as invasive organisms in marine ranches, which were previously done
mainly manually. Therefore, the autonomous monitoring and accurate identification of the
seafood in marine ranches not only helps farmers to control the growth status of seafood
and the habitat changes in real time but also releases manpower from dangerous and
heavy workloads. Numerous advanced acoustic or optical-based detection tools have been
applied to the identification of marine organisms [
1
], and meanwhile underwater robots
are also further considered to provide a larger-scale observation based on their ability to be
incorporated into autonomous mobile devices [2].
In contrast to the acoustic-based approach, the optic-based approach has the advan-
tages of high resolution, low cost, and ease of operation. Thus, optic-based marine organism
identification methods have attracted increasing attention and are becoming a research
J. Mar. Sci. Eng. 2022, 10, 1230. https://doi.org/10.3390/jmse10091230 https://www.mdpi.com/journal/jmse
