CSPPartial-YOLO: A Lightweight YOLO-Based Method for Typical Objects Detection in Remote Sensing Images
Detecting and recognizing objects are crucial steps in interpreting remote sensing images. At present, deep learning methods are predominantly employed for detecting objects in remote sensing images, necessitating a significant number of floating-point computations. However, low computing power and...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10304205/ |
| _version_ | 1797448899403710464 |
|---|---|
| author | Siyu Xie Mei Zhou Chunle Wang Shisheng Huang |
| author_facet | Siyu Xie Mei Zhou Chunle Wang Shisheng Huang |
| author_sort | Siyu Xie |
| collection | DOAJ |
| description | Detecting and recognizing objects are crucial steps in interpreting remote sensing images. At present, deep learning methods are predominantly employed for detecting objects in remote sensing images, necessitating a significant number of floating-point computations. However, low computing power and small storage in computing devices are hard to afford the large model parameter quantity and high computing complexity. To address these constraints, this article presents a lightweight detection model called CSPPartial-YOLO. This model introduces the partial hybrid dilated convolution (PHDC) Block module that combines hybrid dilated convolutions and partial convolutions to increase the receptive field at a low computational cost. By using the PHDC Block within the model design framework of cross-stage partial connection, we construct CSPPartialStage that reduces computational burden without compromising accuracy. Coordinate attention module is also employed in CSPPartialStage to aggregate position information and improve the detection of small objects with complex distributions in remote sensing images. A backbone and neck are developed with CSPPartialStage, and the rotation head of the PPYOLOE-R model adapts to objects of multiple orientations in remote sensing images. Empirical experiments using the dataset for object deTection in aerial images (DOTA) dataset and a large-scale small object detection dAtaset (SODA-A) dataset indicate that our method is faster and resource efficient than the baseline model (PPYOLOE-R), while achieving higher accuracy. Furthermore, comparisons with current state-of-the-art YOLO series detectors show our proposed model's competitiveness in terms of speed and accuracy. Moreover, compared to mainstream lightweight networks, our model exhibits better hardware adaptability, with lower inference latency and higher detection accuracy. |
| first_indexed | 2024-03-09T14:17:06Z |
| format | Article |
| id | doaj.art-c010e3f868814f878a87363006f18d83 |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-09T14:17:06Z |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-c010e3f868814f878a87363006f18d832023-11-29T00:00:59ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352024-01-011738839910.1109/JSTARS.2023.332923510304205CSPPartial-YOLO: A Lightweight YOLO-Based Method for Typical Objects Detection in Remote Sensing ImagesSiyu Xie0https://orcid.org/0009-0006-3894-448XMei Zhou1https://orcid.org/0000-0002-9207-3914Chunle Wang2https://orcid.org/0009-0004-2612-3364Shisheng Huang3Department of Space Microwave Remote Sensing System, Aerospace Information Research Institute, Chinese Academy of Science, Beijing, ChinaDepartment of Space Microwave Remote Sensing System, Aerospace Information Research Institute, Chinese Academy of Science, Beijing, ChinaDepartment of Space Microwave Remote Sensing System, Aerospace Information Research Institute, Chinese Academy of Science, Beijing, ChinaBeijing Institute of Tracking and Telecommunications Technology, Beijing, ChinaDetecting and recognizing objects are crucial steps in interpreting remote sensing images. At present, deep learning methods are predominantly employed for detecting objects in remote sensing images, necessitating a significant number of floating-point computations. However, low computing power and small storage in computing devices are hard to afford the large model parameter quantity and high computing complexity. To address these constraints, this article presents a lightweight detection model called CSPPartial-YOLO. This model introduces the partial hybrid dilated convolution (PHDC) Block module that combines hybrid dilated convolutions and partial convolutions to increase the receptive field at a low computational cost. By using the PHDC Block within the model design framework of cross-stage partial connection, we construct CSPPartialStage that reduces computational burden without compromising accuracy. Coordinate attention module is also employed in CSPPartialStage to aggregate position information and improve the detection of small objects with complex distributions in remote sensing images. A backbone and neck are developed with CSPPartialStage, and the rotation head of the PPYOLOE-R model adapts to objects of multiple orientations in remote sensing images. Empirical experiments using the dataset for object deTection in aerial images (DOTA) dataset and a large-scale small object detection dAtaset (SODA-A) dataset indicate that our method is faster and resource efficient than the baseline model (PPYOLOE-R), while achieving higher accuracy. Furthermore, comparisons with current state-of-the-art YOLO series detectors show our proposed model's competitiveness in terms of speed and accuracy. Moreover, compared to mainstream lightweight networks, our model exhibits better hardware adaptability, with lower inference latency and higher detection accuracy.https://ieeexplore.ieee.org/document/10304205/Deep learningobject detectionpartial convolutionremote sensing image |
| spellingShingle | Siyu Xie Mei Zhou Chunle Wang Shisheng Huang CSPPartial-YOLO: A Lightweight YOLO-Based Method for Typical Objects Detection in Remote Sensing Images IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Deep learning object detection partial convolution remote sensing image |
| title | CSPPartial-YOLO: A Lightweight YOLO-Based Method for Typical Objects Detection in Remote Sensing Images |
| title_full | CSPPartial-YOLO: A Lightweight YOLO-Based Method for Typical Objects Detection in Remote Sensing Images |
| title_fullStr | CSPPartial-YOLO: A Lightweight YOLO-Based Method for Typical Objects Detection in Remote Sensing Images |
| title_full_unstemmed | CSPPartial-YOLO: A Lightweight YOLO-Based Method for Typical Objects Detection in Remote Sensing Images |
| title_short | CSPPartial-YOLO: A Lightweight YOLO-Based Method for Typical Objects Detection in Remote Sensing Images |
| title_sort | csppartial yolo a lightweight yolo based method for typical objects detection in remote sensing images |
| topic | Deep learning object detection partial convolution remote sensing image |
| url | https://ieeexplore.ieee.org/document/10304205/ |
| work_keys_str_mv | AT siyuxie csppartialyoloalightweightyolobasedmethodfortypicalobjectsdetectioninremotesensingimages AT meizhou csppartialyoloalightweightyolobasedmethodfortypicalobjectsdetectioninremotesensingimages AT chunlewang csppartialyoloalightweightyolobasedmethodfortypicalobjectsdetectioninremotesensingimages AT shishenghuang csppartialyoloalightweightyolobasedmethodfortypicalobjectsdetectioninremotesensingimages |