A Feature-Enhanced Small Object Detection Algorithm Based on Attention Mechanism
With the rapid development of AI algorithms and computational power, object recognition based on deep learning frameworks has become a major research direction in computer vision. UAVs equipped with object detection systems are increasingly used in fields like smart transportation, disaster warning,...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/2/589 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832587543107862528 |
|---|---|
| author | Zhe Quan Jun Sun |
| author_facet | Zhe Quan Jun Sun |
| author_sort | Zhe Quan |
| collection | DOAJ |
| description | With the rapid development of AI algorithms and computational power, object recognition based on deep learning frameworks has become a major research direction in computer vision. UAVs equipped with object detection systems are increasingly used in fields like smart transportation, disaster warning, and emergency rescue. However, due to factors such as the environment, lighting, altitude, and angle, UAV images face challenges like small object sizes, high object density, and significant background interference, making object detection tasks difficult. To address these issues, we use YOLOv8s as the basic framework and introduce a multi-level feature fusion algorithm. Additionally, we design an attention mechanism that links distant pixels to improve small object feature extraction. To address missed detections and inaccurate localization, we replace the detection head with a dynamic head, allowing the model to route objects to the appropriate head for final output. We also introduce Slideloss to improve the model’s learning of difficult samples and ShapeIoU to better account for the shape and scale of bounding boxes. Experiments on datasets like VisDrone2019 show that our method improves accuracy by nearly 10% and recall by about 11% compared to the baseline. Additionally, on the AI-TODv1.5 dataset, our method improves the mAP50 from 38.8 to 45.2. |
| format | Article |
| id | doaj-art-ffe815be293c4221a51fe267e7d1f56d |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-ffe815be293c4221a51fe267e7d1f56d2025-01-24T13:49:28ZengMDPI AGSensors1424-82202025-01-0125258910.3390/s25020589A Feature-Enhanced Small Object Detection Algorithm Based on Attention MechanismZhe Quan0Jun Sun1School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi 214122, ChinaSchool of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi 214122, ChinaWith the rapid development of AI algorithms and computational power, object recognition based on deep learning frameworks has become a major research direction in computer vision. UAVs equipped with object detection systems are increasingly used in fields like smart transportation, disaster warning, and emergency rescue. However, due to factors such as the environment, lighting, altitude, and angle, UAV images face challenges like small object sizes, high object density, and significant background interference, making object detection tasks difficult. To address these issues, we use YOLOv8s as the basic framework and introduce a multi-level feature fusion algorithm. Additionally, we design an attention mechanism that links distant pixels to improve small object feature extraction. To address missed detections and inaccurate localization, we replace the detection head with a dynamic head, allowing the model to route objects to the appropriate head for final output. We also introduce Slideloss to improve the model’s learning of difficult samples and ShapeIoU to better account for the shape and scale of bounding boxes. Experiments on datasets like VisDrone2019 show that our method improves accuracy by nearly 10% and recall by about 11% compared to the baseline. Additionally, on the AI-TODv1.5 dataset, our method improves the mAP50 from 38.8 to 45.2.https://www.mdpi.com/1424-8220/25/2/589small object detectionattention mechanismfeature pyramid networkdetection headloss function |
| spellingShingle | Zhe Quan Jun Sun A Feature-Enhanced Small Object Detection Algorithm Based on Attention Mechanism Sensors small object detection attention mechanism feature pyramid network detection head loss function |
| title | A Feature-Enhanced Small Object Detection Algorithm Based on Attention Mechanism |
| title_full | A Feature-Enhanced Small Object Detection Algorithm Based on Attention Mechanism |
| title_fullStr | A Feature-Enhanced Small Object Detection Algorithm Based on Attention Mechanism |
| title_full_unstemmed | A Feature-Enhanced Small Object Detection Algorithm Based on Attention Mechanism |
| title_short | A Feature-Enhanced Small Object Detection Algorithm Based on Attention Mechanism |
| title_sort | feature enhanced small object detection algorithm based on attention mechanism |
| topic | small object detection attention mechanism feature pyramid network detection head loss function |
| url | https://www.mdpi.com/1424-8220/25/2/589 |
| work_keys_str_mv | AT zhequan afeatureenhancedsmallobjectdetectionalgorithmbasedonattentionmechanism AT junsun afeatureenhancedsmallobjectdetectionalgorithmbasedonattentionmechanism AT zhequan featureenhancedsmallobjectdetectionalgorithmbasedonattentionmechanism AT junsun featureenhancedsmallobjectdetectionalgorithmbasedonattentionmechanism |