L-ENet: An Ultralightweight SAR Image Detection Network
Open Access
- 22 April 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
- Vol. 17 (19391404) , 8967-8978
Abstract
Synthetic Aperture Radar (SAR) imagery plays a vital role in maritime vessel detection, despite facing challenges like complex marine environments, low-contrast target recognition, and similarities between ships and sea surface waves. Additionally, the deployment of traditional large models on radar-equipped edge devices is hampered by their extensive computational demands. Addressing these challenges, this study presents an ultra-lightweight network Lightweight Enhanced Network(L-ENet), improving upon YOLOv5n. The backbone network is switched to a more efficient ShufflenetV2, integrated with an improved attention mechanism, DCFAM (Dual-Core Fusion Attention Mechanism), for effective inter-channel information balancing. In the network's neck, the original C3 module has been replaced with the C3EGhost convolution. This convolution module integrates the Ghost and ECA (Efficient Channel Attention) attention mechanisms, aiming to mitigate potential accuracy loss during the lightweight process. Furthermore, this study introduces a novel multi-scale fusion pathway and a Concat module with adaptive weights to better harmonize semantic and detail information. Further, the tri-head detection structure is revised to a dual-head structure OMNI-DIMENSIONAL Adaptive Spatial Feature Fusion (ODASFF), using ODConv to allocate weights across different scales for efficient detection. Experimental results show that L-ENet has a computational cost of 0.6M and a parameter count of 2.1GFLOPs. These figures are 65% and 50% lower than those of the original YOLOv5n while maintaining the same detection accuracy of 97.8%. Hence, L-ENet offers an efficient and viable solution for maritime target detection, showcasing commendable detection performance in comparison with advanced models.Keywords
Funding Information
- National Natural Science Foundation of China (62074017, U2341223)
This publication has 43 references indexed in Scilit:
- DS-YOLOv8-Based Object Detection Method for Remote Sensing ImagesIEEE Access, 2023
- YOLO-UAV: Object Detection Method of Unmanned Aerial Vehicle Imagery Based on Efficient Multi-Scale Feature FusionIEEE Access, 2023
- Scaled-YOLOv4: Scaling Cross Stage Partial Network2021
- GhostNet: More Features From Cheap Operations2020
- ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design2018
- Urban land-use mapping using a deep convolutional neural network with high spatial resolution multispectral remote sensing imageryRemote Sensing of Environment, 2018
- YOLO9000: Better, Faster, Stronger2017
- SSD: Single Shot MultiBox Detector2016
- You Only Look Once: Unified, Real-Time Object Detection2016
- Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation2014