Scholarlink

<p>We present a scene-specific MobileNetV3-DETR for chip-leg defect inspection, where targets are tiny, dense, and arranged on a regular lattice. Instead of stacking generic tricks, we formalize three complementary mechanisms: (A1) a task-aligned FPN pruning criterion that selects the minimal pyramid levels {P3–P5} to match the empirical defect-size distribution; (A2) a lattice-aware relative positional encoding that biases attention toward physically plausible row/column offsets; and (A3) a defect-prior adaptive Top-<i>K</i> sparse attention that allocates the decoder’s attention budget by local response quantiles with a device-aware cap <i>K</i><sub>max</sub>. In training, a defect-aware assignment (D-AA) re-weights the Hungarian classification term by smoothed class priors, improving recall of rare, safety-critical defects without changing the inference graph. Under a unified embedded protocol on Jetson Nano (TensorRT FP16, 512 × 512, batch = 1), the model uses ~ 7.2&#xa0;M parameters and ~ 29.9 GFLOPs, reaches 89.5% mAP@50 and an average of 55.25% mAP@[0.5:0.95], and delivers ≈ 60 ms per image. Ablations demonstrate independent gains from A1–A3, while confusion-matrix analysis confirms reduced errors on the hardest pairs (e.g., bent_leg vs. damaged_leg; dirty vs. scratches), indicating an improved accuracy–latency balance for inline AOI deployment.</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

MobileNetV3-DETR with locality-biased decoding for chip-leg defects on embedded AOI

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月10日

查看全文
  • Zhimin Zhang,
  • Hongxia Liu,
  • Baozhi Huang

摘要

We present a scene-specific MobileNetV3-DETR for chip-leg defect inspection, where targets are tiny, dense, and arranged on a regular lattice. Instead of stacking generic tricks, we formalize three complementary mechanisms: (A1) a task-aligned FPN pruning criterion that selects the minimal pyramid levels {P3–P5} to match the empirical defect-size distribution; (A2) a lattice-aware relative positional encoding that biases attention toward physically plausible row/column offsets; and (A3) a defect-prior adaptive Top-K sparse attention that allocates the decoder’s attention budget by local response quantiles with a device-aware cap Kmax. In training, a defect-aware assignment (D-AA) re-weights the Hungarian classification term by smoothed class priors, improving recall of rare, safety-critical defects without changing the inference graph. Under a unified embedded protocol on Jetson Nano (TensorRT FP16, 512 × 512, batch = 1), the model uses ~ 7.2 M parameters and ~ 29.9 GFLOPs, reaches 89.5% mAP@50 and an average of 55.25% mAP@[0.5:0.95], and delivers ≈ 60 ms per image. Ablations demonstrate independent gains from A1–A3, while confusion-matrix analysis confirms reduced errors on the hardest pairs (e.g., bent_leg vs. damaged_leg; dirty vs. scratches), indicating an improved accuracy–latency balance for inline AOI deployment.