DERA框架增强X射线图像违禁品检测能力

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-14 04:00

研究人员开发了DERA，一个旨在提高X射线图像中违禁品检测能力的框架。该方法结合了分层视觉特征和像素差边缘金字塔，从训练数据中学习特定对象的边界先验。DERA的架构通过分离边界监督并将可训练参数限制在约14.7K个，从而实现适应性。在PIDray、CLCXray和STCray数据集上进行测试时，DERA分别比基线模型在平均精度上提高了3.1、1.6和2.4个百分点。 AI

影响这项研究通过提高X射线扫描中隐藏物品的检测能力，有望带来更准确、更高效的安全筛查系统。

排序理由该集群是一篇详细介绍新技术框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 31 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群是一篇详细介绍新技术框架的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, other; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CV TIER_1 English(EN) · Yonathan Michael, Mohamad Alansari, Mohammed Bennamoun, Dwarikanath Mahapatra, Andreas Henschel, Naoufel Werghi · 2026-09-14 04:00

DERA：分离式边缘残差适配用于违禁品检测

arXiv:2609.12411v1 Announce Type: new Abstract: Prohibited-item detection in X-ray imagery remains challenging due to object superposition, weak texture, and material clutter which obscure both semantic appearance and object boundaries. We propose \textbf{DERA}, a \textbf{D}etach…

本文如何被排名

报道来源 [1]

DERA：分离式边缘残差适配用于违禁品检测

相关实体

相关话题