Unsupervised Infrared Small-Object-Detection Approach of Spatial–Temporal Patch Tensor and Object Selection
In this study, an unsupervised infrared object-detection approach based on spatial–temporal patch tensor and object selection is proposed to fully use effective temporal information and maintain a balance between object-detection performance and computation time. Initially, a spatial–temporal patch...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-03-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/14/7/1612 |
_version_ | 1797437862178717696 |
---|---|
author | Ruixi Zhu Long Zhuang |
author_facet | Ruixi Zhu Long Zhuang |
author_sort | Ruixi Zhu |
collection | DOAJ |
description | In this study, an unsupervised infrared object-detection approach based on spatial–temporal patch tensor and object selection is proposed to fully use effective temporal information and maintain a balance between object-detection performance and computation time. Initially, a spatial–temporal patch tensor is proposed by performing median pooling function on patch tensors generated from consecutive frames to suppress sky or cloud clutter. Then, a contrast-boosted approach that incorporates morphological operations is proposed to improve the contrast between objects and background. Finally, an object-selection approach is proposed based on the cluster center derived from clustering locations and gray values, thereby decreasing the search scope of objects in the detection process. The experiments of five infrared sequence frames confirm that the proposed framework can obtain better results than most previous methods when handling heterogeneous scenes in terms of gray values. Experimental results of five real sequence frames also demonstrate that the spatial–temporal patch tensor, the contrast-boosted approach, and object-selection approach can increase the recall ratio by 6.7, 2.21, and 1.14 percentage units and the precision ratio by 1.61, 3.44, and 11.79 percentage units, respectively. Moreover, the proposed framework can achieve an average F1 score of 0.9804 with about 1.85 s of computation time, demonstrating that it can obtain satisfactory object-detection performance with relatively low computation time. |
first_indexed | 2024-03-09T11:28:45Z |
format | Article |
id | doaj.art-b3f17506aae949a784a34cf89923e1ed |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-09T11:28:45Z |
publishDate | 2022-03-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-b3f17506aae949a784a34cf89923e1ed2023-11-30T23:56:29ZengMDPI AGRemote Sensing2072-42922022-03-01147161210.3390/rs14071612Unsupervised Infrared Small-Object-Detection Approach of Spatial–Temporal Patch Tensor and Object SelectionRuixi Zhu0Long Zhuang1Department of Research, Nanjing Research Institute of Electronic Technology, Nanjing 210039, ChinaDepartment of Research, Nanjing Research Institute of Electronic Technology, Nanjing 210039, ChinaIn this study, an unsupervised infrared object-detection approach based on spatial–temporal patch tensor and object selection is proposed to fully use effective temporal information and maintain a balance between object-detection performance and computation time. Initially, a spatial–temporal patch tensor is proposed by performing median pooling function on patch tensors generated from consecutive frames to suppress sky or cloud clutter. Then, a contrast-boosted approach that incorporates morphological operations is proposed to improve the contrast between objects and background. Finally, an object-selection approach is proposed based on the cluster center derived from clustering locations and gray values, thereby decreasing the search scope of objects in the detection process. The experiments of five infrared sequence frames confirm that the proposed framework can obtain better results than most previous methods when handling heterogeneous scenes in terms of gray values. Experimental results of five real sequence frames also demonstrate that the spatial–temporal patch tensor, the contrast-boosted approach, and object-selection approach can increase the recall ratio by 6.7, 2.21, and 1.14 percentage units and the precision ratio by 1.61, 3.44, and 11.79 percentage units, respectively. Moreover, the proposed framework can achieve an average F1 score of 0.9804 with about 1.85 s of computation time, demonstrating that it can obtain satisfactory object-detection performance with relatively low computation time.https://www.mdpi.com/2072-4292/14/7/1612unsupervised infrared object detectionspatial–temporal patch tensorcontrast-boosted approachobject-selection approachreal infrared image sequences |
spellingShingle | Ruixi Zhu Long Zhuang Unsupervised Infrared Small-Object-Detection Approach of Spatial–Temporal Patch Tensor and Object Selection Remote Sensing unsupervised infrared object detection spatial–temporal patch tensor contrast-boosted approach object-selection approach real infrared image sequences |
title | Unsupervised Infrared Small-Object-Detection Approach of Spatial–Temporal Patch Tensor and Object Selection |
title_full | Unsupervised Infrared Small-Object-Detection Approach of Spatial–Temporal Patch Tensor and Object Selection |
title_fullStr | Unsupervised Infrared Small-Object-Detection Approach of Spatial–Temporal Patch Tensor and Object Selection |
title_full_unstemmed | Unsupervised Infrared Small-Object-Detection Approach of Spatial–Temporal Patch Tensor and Object Selection |
title_short | Unsupervised Infrared Small-Object-Detection Approach of Spatial–Temporal Patch Tensor and Object Selection |
title_sort | unsupervised infrared small object detection approach of spatial temporal patch tensor and object selection |
topic | unsupervised infrared object detection spatial–temporal patch tensor contrast-boosted approach object-selection approach real infrared image sequences |
url | https://www.mdpi.com/2072-4292/14/7/1612 |
work_keys_str_mv | AT ruixizhu unsupervisedinfraredsmallobjectdetectionapproachofspatialtemporalpatchtensorandobjectselection AT longzhuang unsupervisedinfraredsmallobjectdetectionapproachofspatialtemporalpatchtensorandobjectselection |