Scholarlink

<p>Handwritten text-line segmentation in historical manuscripts remains challenging due to degradation, overlapping strokes, and extreme data scarcity. We propose APAU-Net, a two-stage cascade architecture that improves Line Intersection-over-Union (Line IU) metric via learned anisotropic Gaussian priors. Stage 1 predicts topology-aware ellipsoidal priors from low resolution grayscale images using connected-component analysis and moment-based ellipse fitting. Stage 2 refines these priors at full resolution through a residual U-Net with adaptive per-pixel weighting and gated fusion. We evaluated APAU-Net on the challenging U-DIADS-TL (84 images, only 3 training pages per manuscript) and DIVA-HisDB benchmarks. It achieves an average Line Intersection-over-Union (Line IU) of 94.3%, corresponding to a 9.83 percentage-point improvement over plain U-Net, and improves performance by 24.08 percentage points over the strongest few-shot baseline on the most degraded Syriac subset. Ablation confirms that the anisotropic prior contributes 8.61–11.67 percentage points to Line IU under severe data scarcity. The source code is available at https://github.com/aminebeg/APAUNet.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

APAU-Net: Adaptive prior-aware U-Net for handwritten text-line segmentation in historical documents

  • 期刊论文

  • 发表日期：

    2026年6月30日

查看全文
  • Mohamed Amine Beghoura,
  • Abdelouahab Attia,
  • Abderraouf Bouziane,
  • M. Hassaballah

摘要

Handwritten text-line segmentation in historical manuscripts remains challenging due to degradation, overlapping strokes, and extreme data scarcity. We propose APAU-Net, a two-stage cascade architecture that improves Line Intersection-over-Union (Line IU) metric via learned anisotropic Gaussian priors. Stage 1 predicts topology-aware ellipsoidal priors from low resolution grayscale images using connected-component analysis and moment-based ellipse fitting. Stage 2 refines these priors at full resolution through a residual U-Net with adaptive per-pixel weighting and gated fusion. We evaluated APAU-Net on the challenging U-DIADS-TL (84 images, only 3 training pages per manuscript) and DIVA-HisDB benchmarks. It achieves an average Line Intersection-over-Union (Line IU) of 94.3%, corresponding to a 9.83 percentage-point improvement over plain U-Net, and improves performance by 24.08 percentage points over the strongest few-shot baseline on the most degraded Syriac subset. Ablation confirms that the anisotropic prior contributes 8.61–11.67 percentage points to Line IU under severe data scarcity. The source code is available at https://github.com/aminebeg/APAUNet.