Abstract:A lightweight underwater fish image semantic segmentation model is proposed based on U‐Net. The multi‐input mode, MRS module, MSC structure module and CBAM attention mechanism are introduced into U‐Net network. The results show that the proposed model has the advantages of high accuracy and a small amount of parameters, and achieves the balance between accuracy and speed, which is more suitable for underwater image segmentation. Semantic segmentation of underwater fish images is vital for monitoring fish stocks, assessing marine resources, and sustaining fisheries. To tackle challenges such as low segmentation accuracy, inadequate real‐time performance, and imprecise location segmentation in current methods, a novel lightweight U‐Net model is proposed. The proposed model acquires more segmentation details by applying a multiple‐input approach at the first four encoder levels. To achieve both lightweight and high accuracy, a multi‐scale residual structure (MRS) module is proposed to reduce parameters and compensate for the accuracy loss caused by the reduction of channels. To improve segmentation accuracy, a multi‐scale skip connection (MSC) structure is further proposed, and the convolution block attention mechanism (CBAM) is introduced at the end of each decoder level for weight adjustment. Experimental results demonstrate a notable reduction in model volume, parameters, and floating‐point operations by 94.20%, 94.39%, and 51.52% respectively, compared to the original model. The proposed model achieves a high mean intersection over union (mIOU) of 94.44%, mean pixel accuracy (mPA) of 97.03%, and a frame rate of 43.62 frames per second (FPS). With its high precision and minimal parameters, the model strikes a balance between accuracy and speed, making it particularly suitable for underwater image segmentation.

Image Semantic Segmentation of Underwater Garbage with Modified U-Net Architecture Model

Real-time Water Area Segmentation for USV using Enhanced U-Net

Depth-Conditioned GAN for Underwater Image Enhancement

Lightweight deep learning model for underwater waste segmentation based on sonar images

WaterBiSeg-Net: An underwater bilateral segmentation network for marine debris segmentation

An Underwater Image Semantic Segmentation Method Focusing on Boundaries and a Real Underwater Scene Semantic Segmentation Dataset

Garbage Detection on The Water Surface Based on Deep Learning

An Embeddable Algorithm for Automatic Garbage Detection Based on Complex Marine Environment

An Underwater Image Enhancement Model for Domain Adaptation

Instance Segmentation of Underwater Images by Using Deep Learning

A lightweight underwater fish image semantic segmentation model based on U‐Net

Full stage networks with auxiliary focal loss and multi-attention module for submarine garbage object detection

Enhanced YOLOv8-Seg Instance Segmentation for Real-Time Submerged Debris Detection

Semantic Segmentation of Fish and Underwater Environments Using Deep Convolutional Neural Networks and Learned Active Contours

YOLOv8-C2f-Faster-EMA: An Improved Underwater Trash Detection Model Based on YOLOv8

A modified YOLOv4 detection method for a vision-based underwater garbage cleaning robot

Analysis of Deep Learning Based Garbage Detection in Water Bodies

Lightweight Underwater Object Detection Algorithm for Embedded Deployment Using Higher-Order Information and Image Enhancement

A modified YOLOv3 detection method for vision-based water surface garbage capture robot

A Dataset for Detection and Segmentation of Underwater Marine Debris in Shallow Waters

Semantic Segmentation of Underwater Imagery: Dataset and Benchmark