Alapfogalmak
Proposing a diffusion model framework for IRSTD to address target-level insensitivity and designing a low-frequency isolation module in the wavelet domain.
Kivonat
The content discusses the challenges of detecting small targets in infrared clutter background and proposes a generative approach, IRSTD-Diff, to overcome target-level insensitivity. The diffusion model compensates for false alarms and missed detections by modeling mask posterior distribution. A low-frequency isolation module is designed to reduce interference from low-level noise. Experimental results show superior performance over state-of-the-art methods on three datasets.
- Introduction to Infrared Small Target Detection (IRSTD)
- Challenges faced in IRSTD due to small target size and clutter background
- Proposal of IRSTD-Diff using a diffusion model framework and low-frequency isolation module
- Explanation of how IRSTD-Diff addresses target-level insensitivity and reduces false alarms and missed detections
- Overview of experimental results showing superior performance on three datasets compared to existing methods
Statisztikák
提案されたIRSTD-Diffは、偽の警告と見逃し検出を補償するためにマスク事後分布をモデル化します。
低周波数分離モジュールは、低レベルのノイズからの干渉を軽減するために設計されています。
IRSTD-Diffは既存の方法に比べて3つのデータセットで優れたパフォーマンスを示しています。