Water‐surface infrared small object detection based on spatial feature weighting and class balancing method
Abstract Infrared imaging is widely used due to its penetration capability to operate under many weather or lighting condition. However, due to the far distance of aerial view, feature blur, and the scarcity of aerial infrared data, the detection of small infrared targets on the water surface remain...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2023-08-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.12851 |
| _version_ | 1797755446537224192 |
|---|---|
| author | Tian Hui YueLei Xu HuaFeng Li Qing Zhou Jarhinbek Rasol |
| author_facet | Tian Hui YueLei Xu HuaFeng Li Qing Zhou Jarhinbek Rasol |
| author_sort | Tian Hui |
| collection | DOAJ |
| description | Abstract Infrared imaging is widely used due to its penetration capability to operate under many weather or lighting condition. However, due to the far distance of aerial view, feature blur, and the scarcity of aerial infrared data, the detection of small infrared targets on the water surface remains a challenging problem. In response to the problem of unclear features, we propose the spatial feature weighting method based on 2D Gaussian distribution. This method increases the weight of the target area by adaptively adjusting the feature activation. Secondly, for the problem of rare aerial perspective infrared data, we propose the cross‐spectral data migration method. By introducing the domain difference loss function to optimize the pseudo‐label selection process, the range of target domain distribution is expanded, and the adaptability of the detector is improved. Finally, in response to the problem of underfitting caused by category imbalance in transfer learning, we propose the class balancing method that effectively reduces the false detection. Extensive experiments were conducted on both benchmark datasets and the self‐built dataset to evaluate the effectiveness and robustness of our method. The proposed method was evaluated with different models and various scenarios, and the results demonstrated the effectiveness. |
| first_indexed | 2024-03-12T17:48:00Z |
| format | Article |
| id | doaj.art-d9a51b3d6cfd4ecaa5af91fec827a981 |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-03-12T17:48:00Z |
| publishDate | 2023-08-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-d9a51b3d6cfd4ecaa5af91fec827a9812023-08-03T12:43:17ZengWileyIET Image Processing1751-96591751-96672023-08-0117103012302710.1049/ipr2.12851Water‐surface infrared small object detection based on spatial feature weighting and class balancing methodTian Hui0YueLei Xu1HuaFeng Li2Qing Zhou3Jarhinbek Rasol4Institute of Unmanned System Research Northwestern Polytechnical University Xi'anChinaInstitute of Unmanned System Research Northwestern Polytechnical University Xi'anChinaInstitute of Unmanned System Research Northwestern Polytechnical University Xi'anChinaInstitute of Unmanned System Research Northwestern Polytechnical University Xi'anChinaInstitute of Unmanned System Research Northwestern Polytechnical University Xi'anChinaAbstract Infrared imaging is widely used due to its penetration capability to operate under many weather or lighting condition. However, due to the far distance of aerial view, feature blur, and the scarcity of aerial infrared data, the detection of small infrared targets on the water surface remains a challenging problem. In response to the problem of unclear features, we propose the spatial feature weighting method based on 2D Gaussian distribution. This method increases the weight of the target area by adaptively adjusting the feature activation. Secondly, for the problem of rare aerial perspective infrared data, we propose the cross‐spectral data migration method. By introducing the domain difference loss function to optimize the pseudo‐label selection process, the range of target domain distribution is expanded, and the adaptability of the detector is improved. Finally, in response to the problem of underfitting caused by category imbalance in transfer learning, we propose the class balancing method that effectively reduces the false detection. Extensive experiments were conducted on both benchmark datasets and the self‐built dataset to evaluate the effectiveness and robustness of our method. The proposed method was evaluated with different models and various scenarios, and the results demonstrated the effectiveness.https://doi.org/10.1049/ipr2.12851computer visionconvolutional neural netsfeature extractionGaussian distributionimage processinginfrared detectors |
| spellingShingle | Tian Hui YueLei Xu HuaFeng Li Qing Zhou Jarhinbek Rasol Water‐surface infrared small object detection based on spatial feature weighting and class balancing method IET Image Processing computer vision convolutional neural nets feature extraction Gaussian distribution image processing infrared detectors |
| title | Water‐surface infrared small object detection based on spatial feature weighting and class balancing method |
| title_full | Water‐surface infrared small object detection based on spatial feature weighting and class balancing method |
| title_fullStr | Water‐surface infrared small object detection based on spatial feature weighting and class balancing method |
| title_full_unstemmed | Water‐surface infrared small object detection based on spatial feature weighting and class balancing method |
| title_short | Water‐surface infrared small object detection based on spatial feature weighting and class balancing method |
| title_sort | water surface infrared small object detection based on spatial feature weighting and class balancing method |
| topic | computer vision convolutional neural nets feature extraction Gaussian distribution image processing infrared detectors |
| url | https://doi.org/10.1049/ipr2.12851 |
| work_keys_str_mv | AT tianhui watersurfaceinfraredsmallobjectdetectionbasedonspatialfeatureweightingandclassbalancingmethod AT yueleixu watersurfaceinfraredsmallobjectdetectionbasedonspatialfeatureweightingandclassbalancingmethod AT huafengli watersurfaceinfraredsmallobjectdetectionbasedonspatialfeatureweightingandclassbalancingmethod AT qingzhou watersurfaceinfraredsmallobjectdetectionbasedonspatialfeatureweightingandclassbalancingmethod AT jarhinbekrasol watersurfaceinfraredsmallobjectdetectionbasedonspatialfeatureweightingandclassbalancingmethod |