YOLOv6-ESG: A Lightweight Seafood Detection Method

Abstract

The rapid development of convolutional neural networks has significant implications for automated underwater fishing operations. Among these, object detection algorithms based on underwater robots have become a hot topic in both academic and applied research. Due to the complexity of underwater imaging environments, many studies have employed large network structures to enhance the model’s detection accuracy. However, such models contain many parameters and consume substantial memory, making them less suitable for small devices with limited memory and computing capabilities. To address these issues, a YOLOv6-based lightweight underwater object detection model, YOLOv6-ESG, is proposed to detect seafood, such as echinus, holothurian, starfish, and scallop. First, a more lightweight backbone network is designed by rebuilding the EfficientNetv2 with a lightweight ODConv module to reduce the number of parameters and floating-point operations. Then, this study improves the neck layer using lightweight GSConv and VoVGSCSP modules to enhance the network’s ability to detect small objects. Meanwhile, to improve the detection accuracy of small underwater objects with poor image quality and low resolution, the SPD-Conv module is also integrated into the two parts of the model. Finally, the Adan optimizer is utilized to speed up model convergence and further improve detection accuracy. To address the issue of interference objects in the URPC2022 dataset, data cleaning has been conducted, followed by experiments on the cleaned dataset. The proposed model achieves 86.6% mAP while the detection speed (batch size = 1) reaches 50.66 FPS. Compared to YOLOv6, the proposed model not only maintains almost the same level of detection accuracy but also achieves faster detection speed. Moreover, the number of parameters and floating-point operations reaches the minimum levels, with reductions of 75.44% and 79.64%, respectively. These results indicate the feasibility of the proposed model in the application of underwater detection tasks.

A Lightweight Model for Real-Time Detection of Vehicle Black Smoke

“…The GSConv module enhances the network's non-linear expressive capability by adding depthwise separable convolution layers and channel shuffling operations. Experiments have shown that introducing the GSConv module to replace regular Conv in various stages of the model results in a deeper network [34]. When the feature maps extracted by the YOLOv5s model's backbone network are input to the neck network, their channel size has reached its maximum value, while the width and height dimensions are at their minimum.…”

YOLOv7t-CEBC Network for Underwater Litter Detection

“…At the same time, the latter, exemplified by SSD [9] (single-shot multi-box detectors; region-based detection strategies) and the YOLO [10] (You Only Look Once: a widely used model known for its speed and precision; it was first introduced by Joseph Redmon et al in 2016 and has since undergone several iterations) series of algorithms, where we directly extracted features and performed object classification and localization using CNNs. This type of algorithm has been widely embraced in object detection due to its faster detection speed [11].…”

Deploying a Computer Vision Model Based on YOLOv8 Suitable for Drones in the Tuna Fishing and Aquaculture Industry

“…Concurrently, the elimination of superfluous prediction layers streamlines the model, tailoring it more effectively to the identification of smaller marine species. The integration of state-of-the-art techniques such as the Global Attention Module (GAM) [12][13][14] and Speed-up Convolution (SPD-Conv) [15][16][17][18] further fortifies the model's capacity for learning and stabilization, guaranteeing an exceptional performance standard in the detection and analysis of information pertaining to small or distant marine entities.…”