摘要
Millimeter-wave (MMW) human screening offers non-ionizing operation with limited penetration through clothing, yet reliable contraband detection remains challenging. Low signal-to-noise ratios, complex scattering, substantial pose variability, and weak target-background separability often render prohibited items small, faint, and easily confounded. This paper reviews recent advances in deep learning for MMW threat detection and alarm generation, categorized into four complementary research directions: (i) quality-driven detectability improvements; (ii) detector optimization for small object recognition; (iii) incorporation of human-structure priors for false-alarm suppression; and (iv) multi-view/multi-source fusion for robust screening. Furthermore, we discuss deployment-critical issues including trustworthy alarms, real-time inference, and domain generalization, and outline promising avenues for future research.
| 源语言 | 英语 |
|---|---|
| 主期刊名 | 2026 International Conference on Microwave and Millimeter Wave Technology, ICMMT 2026 - Proceedings |
| 出版商 | Institute of Electrical and Electronics Engineers Inc. |
| 版本 | 2026 |
| ISBN(电子版) | 9798319509628 |
| DOI | |
| 出版状态 | 已出版 - 2026 |
| 已对外发布 | 是 |
| 活动 | 17th International Conference on Microwave and Millimeter Wave Technology, ICMMT 2026 - Shenzhen, 中国 期限: 9 5月 2026 → 12 5月 2026 |
会议
| 会议 | 17th International Conference on Microwave and Millimeter Wave Technology, ICMMT 2026 |
|---|---|
| 国家/地区 | 中国 |
| 市 | Shenzhen |
| 时期 | 9/05/26 → 12/05/26 |
学术指纹
探究 'Toward Reliable MMW Screening: A Survey of Deep Learning for Threat Detection' 的科研主题。它们共同构成独一无二的学术指纹。引用此
- APA
- Author
- BIBTEX
- Harvard
- Standard
- RIS
- Vancouver