Few-shot concealed object detection in sub-THz security images using improved pseudo-annotations
Abstract In this research, we explore the few-shot object detection application for identifying concealed objects in sub-terahertz security images, using fine-tuning based frameworks. To adapt these machine learning frameworks for the (sub-)terahertz domain, we propose an innovative pseudo-annotatio...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2024-02-01
|
| Series: | Scientific Reports |
| Online Access: | https://doi.org/10.1038/s41598-024-53045-9 |
| _version_ | 1797275007373541376 |
|---|---|
| author | Ran Cheng Stepan Lucyszyn |
| author_facet | Ran Cheng Stepan Lucyszyn |
| author_sort | Ran Cheng |
| collection | DOAJ |
| description | Abstract In this research, we explore the few-shot object detection application for identifying concealed objects in sub-terahertz security images, using fine-tuning based frameworks. To adapt these machine learning frameworks for the (sub-)terahertz domain, we propose an innovative pseudo-annotation method to augment the object detector by sourcing high-quality training samples from unlabeled images. This approach employs multiple one-class detectors coupled with a fine-grained classifier, trained on supporting thermal-infrared images, to prevent overfitting. Consequently, our approach enhances the model’s ability to detect challenging objects (e.g., 3D-printed guns and ceramic knives) when few-shot training examples are available, especially in the real-world scenario where images of concealed dangerous items are scarce. |
| first_indexed | 2024-03-07T15:08:10Z |
| format | Article |
| id | doaj.art-6499663dd260436284837cd7543aba4e |
| institution | Directory Open Access Journal |
| issn | 2045-2322 |
| language | English |
| last_indexed | 2024-03-07T15:08:10Z |
| publishDate | 2024-02-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj.art-6499663dd260436284837cd7543aba4e2024-03-05T18:49:26ZengNature PortfolioScientific Reports2045-23222024-02-011411910.1038/s41598-024-53045-9Few-shot concealed object detection in sub-THz security images using improved pseudo-annotationsRan Cheng0Stepan Lucyszyn1Department of Electrical and Electronic Engineering, Imperial College LondonDepartment of Electrical and Electronic Engineering, Imperial College LondonAbstract In this research, we explore the few-shot object detection application for identifying concealed objects in sub-terahertz security images, using fine-tuning based frameworks. To adapt these machine learning frameworks for the (sub-)terahertz domain, we propose an innovative pseudo-annotation method to augment the object detector by sourcing high-quality training samples from unlabeled images. This approach employs multiple one-class detectors coupled with a fine-grained classifier, trained on supporting thermal-infrared images, to prevent overfitting. Consequently, our approach enhances the model’s ability to detect challenging objects (e.g., 3D-printed guns and ceramic knives) when few-shot training examples are available, especially in the real-world scenario where images of concealed dangerous items are scarce.https://doi.org/10.1038/s41598-024-53045-9 |
| spellingShingle | Ran Cheng Stepan Lucyszyn Few-shot concealed object detection in sub-THz security images using improved pseudo-annotations Scientific Reports |
| title | Few-shot concealed object detection in sub-THz security images using improved pseudo-annotations |
| title_full | Few-shot concealed object detection in sub-THz security images using improved pseudo-annotations |
| title_fullStr | Few-shot concealed object detection in sub-THz security images using improved pseudo-annotations |
| title_full_unstemmed | Few-shot concealed object detection in sub-THz security images using improved pseudo-annotations |
| title_short | Few-shot concealed object detection in sub-THz security images using improved pseudo-annotations |
| title_sort | few shot concealed object detection in sub thz security images using improved pseudo annotations |
| url | https://doi.org/10.1038/s41598-024-53045-9 |
| work_keys_str_mv | AT rancheng fewshotconcealedobjectdetectioninsubthzsecurityimagesusingimprovedpseudoannotations AT stepanlucyszyn fewshotconcealedobjectdetectioninsubthzsecurityimagesusingimprovedpseudoannotations |