Light-YOLO: a lightweight detection algorithm based on multi-scale feature enhancement for infrared small ship target
Abstract Infrared-based detection of small targets on ships is crucial for ensuring navigation safety and effective maritime traffic management. However, existing ship target detection models often encounter missed detections and struggle to achieve both high accuracy and real-time performance at th...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2025-01-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01726-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823861463450648576 |
|---|---|
| author | Ji Tang Xiao-Min Hu Sang-Woon Jeon Wei-Neng Chen |
| author_facet | Ji Tang Xiao-Min Hu Sang-Woon Jeon Wei-Neng Chen |
| author_sort | Ji Tang |
| collection | DOAJ |
| description | Abstract Infrared-based detection of small targets on ships is crucial for ensuring navigation safety and effective maritime traffic management. However, existing ship target detection models often encounter missed detections and struggle to achieve both high accuracy and real-time performance at the same time. Addressing these challenges, this study presents Light-YOLO, a lightweight model for ship small target detection. Within the YOLOv8 network architecture, Light-YOLO replaces conventional convolutions with snake convolutions, effectively addressing the issue of inadequate detection point receptive fields for small targets, thereby enhancing their detection. Additionally, a Multi-Scale Feature Enhancement Module (MFEB) is introduced to refine focus on low-level features through multi-scale and selection strategies, mitigating issues such as interference from image backgrounds and noise during small target detection. Furthermore, a novel loss function is designed to dynamically adjust the proportions of its components during training, improving the regression accuracy of small targets towards real annotation boxes and enhancing the localization ability of detection boxes. Experimental results demonstrate that Light-YOLO outperforms YOLOv8n, achieving optimal performance on an infrared ship small target detection dataset with 9.2G FLOPs. It notably enhances accuracy, recall rate, and average precision by 1.76%, 0.83%, and 2.27%, respectively. |
| format | Article |
| id | doaj-art-4cf37f5446c64c88bcf1b8ecf7ab8170 |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-4cf37f5446c64c88bcf1b8ecf7ab81702025-02-09T13:01:00ZengSpringerComplex & Intelligent Systems2199-45362198-60532025-01-0111211210.1007/s40747-024-01726-3Light-YOLO: a lightweight detection algorithm based on multi-scale feature enhancement for infrared small ship targetJi Tang0Xiao-Min Hu1Sang-Woon Jeon2Wei-Neng Chen3School of Computer Science and Engineering, South China University of TechnologySchool of Computer Science and Technology, Guangdong University of TechnologyDepartment of Electrical and Electronic Engineering, Hanyang UniversitySchool of Computer Science and Engineering, South China University of TechnologyAbstract Infrared-based detection of small targets on ships is crucial for ensuring navigation safety and effective maritime traffic management. However, existing ship target detection models often encounter missed detections and struggle to achieve both high accuracy and real-time performance at the same time. Addressing these challenges, this study presents Light-YOLO, a lightweight model for ship small target detection. Within the YOLOv8 network architecture, Light-YOLO replaces conventional convolutions with snake convolutions, effectively addressing the issue of inadequate detection point receptive fields for small targets, thereby enhancing their detection. Additionally, a Multi-Scale Feature Enhancement Module (MFEB) is introduced to refine focus on low-level features through multi-scale and selection strategies, mitigating issues such as interference from image backgrounds and noise during small target detection. Furthermore, a novel loss function is designed to dynamically adjust the proportions of its components during training, improving the regression accuracy of small targets towards real annotation boxes and enhancing the localization ability of detection boxes. Experimental results demonstrate that Light-YOLO outperforms YOLOv8n, achieving optimal performance on an infrared ship small target detection dataset with 9.2G FLOPs. It notably enhances accuracy, recall rate, and average precision by 1.76%, 0.83%, and 2.27%, respectively.https://doi.org/10.1007/s40747-024-01726-3Infrared small target detectionSnake convolutionMulti-scale feature enhancementYOLOv8 |
| spellingShingle | Ji Tang Xiao-Min Hu Sang-Woon Jeon Wei-Neng Chen Light-YOLO: a lightweight detection algorithm based on multi-scale feature enhancement for infrared small ship target Complex & Intelligent Systems Infrared small target detection Snake convolution Multi-scale feature enhancement YOLOv8 |
| title | Light-YOLO: a lightweight detection algorithm based on multi-scale feature enhancement for infrared small ship target |
| title_full | Light-YOLO: a lightweight detection algorithm based on multi-scale feature enhancement for infrared small ship target |
| title_fullStr | Light-YOLO: a lightweight detection algorithm based on multi-scale feature enhancement for infrared small ship target |
| title_full_unstemmed | Light-YOLO: a lightweight detection algorithm based on multi-scale feature enhancement for infrared small ship target |
| title_short | Light-YOLO: a lightweight detection algorithm based on multi-scale feature enhancement for infrared small ship target |
| title_sort | light yolo a lightweight detection algorithm based on multi scale feature enhancement for infrared small ship target |
| topic | Infrared small target detection Snake convolution Multi-scale feature enhancement YOLOv8 |
| url | https://doi.org/10.1007/s40747-024-01726-3 |
| work_keys_str_mv | AT jitang lightyoloalightweightdetectionalgorithmbasedonmultiscalefeatureenhancementforinfraredsmallshiptarget AT xiaominhu lightyoloalightweightdetectionalgorithmbasedonmultiscalefeatureenhancementforinfraredsmallshiptarget AT sangwoonjeon lightyoloalightweightdetectionalgorithmbasedonmultiscalefeatureenhancementforinfraredsmallshiptarget AT weinengchen lightyoloalightweightdetectionalgorithmbasedonmultiscalefeatureenhancementforinfraredsmallshiptarget |