Aug 2026· Journal of Computing and Electronic Information Management· Vol 22, pp. 38-41· 0 citations· 3 references
TL;DR
The results show that targeted attention can improve moderate-overlap detection without the cost of an additional high-resolution head, but they also expose unresolved localization, source-domain, and false-accept limitations.
Abstract
Small chip-surface defects are difficult to detect because they occupy few pixels and vary across acquisition domains. This study develops a compact YOLOv8n detector through controlled, validation-only ablations while preserving a held-out test set for later model development. The dataset comprised 3,511 real chip images from two source domains and three defect classes. Increasing the input resolution from 416 to 640 pixels improved validation mAP50 from 0.755 to 0.778, with the largest class-level gain for the smallest DIE_INK defects. Adding a stride-4 P2 head increased computation from 8.1 to 12.2 GFLOPs but reduced mAP50 to 0.729. In contrast, placing a convolutional block attention module on the P3 feature map increased mAP50 to 0.797 with a negligible parameter increase, although mAP50-95 decreased slightly from 0.468 to 0.463. Focal loss and a deterministic Dataset-B-style augmentation both degraded the relevant validation metrics. At a validation-selected confidence threshold of 0.15, the detector-supported sorting rule achieved 95.44% three-way decision accuracy and a 5.88% defect false-accept rate. These results show that targeted attention can improve moderate-overlap detection without the cost of an additional high-resolution head, but they also expose unresolved localization, source-domain, and false-accept limitations. The reported values are development-stage validation evidence and are not estimates of final held-out generalization.
The proposed DAF-YOLO model, an enhanced YOLO11 model for detecting small defects with weak visual features on chip surfaces, incorporates a dual-pooling channel-spatial attention module (DPCSA) to strengthen salient responses in small-defect regions and suppress interference from complex background textures.
Kai Wang, Ji-Nan Gu, Juan Huang· Engineering Research Express· 0 citations
Industrial surface inspection must localize defects when annotated data are limited. We present Asymmetric CNN–ViT Residual Fusion (ACVRF), in which a convolutional neural network (CNN) generates proposals and primary region-of-interest features, while a Vision Transformer (ViT) supplies a residual with a bounded mixin...
Chao Ding, Kun Zou, Yu Zhong et al.· Engineer· 0 citations
Printed circuit board (PCB) defect detection remains challenging because micro-scale defects often exhibit weak contrast and are easily suppressed by conventional downsampling and balanced multi-scale detection heads. To address these limitations, this paper proposes YOLO-SAM, a lightweight YOLOv8n-based detector with...
Zhi-Yan Zhong, Bei Zhang, Hong-Xin Wang· Symmetry· 0 citations
Reliable printed circuit board (PCB) inspection must identify defects that occupy only a few pixels while separating categories with highly similar visual patterns. To address the resulting loss of detail and unstable localization, an improved YOLOv8 detector is developed with a high-resolution small-target branch, a r...
Jun Wang, Zhi-Qiang Fu, Yi Wu· Engineering Research Express· 0 citations
Printed circuit board (PCB) defect detection is critical for ensuring the performance and reliability of electronic products. However, traditional inspection methods suffer from low efficiency and poor stability. In recent years, the development of deep learning models has proceeded at a rapid pace. But existing deep l...
Printed product surface defect inspection faces critical challenges in real industrial production, including insufficient detection accuracy, poor robustness, and scarcity of labeled defective samples. This paper proposes an improved Siamese network for high-resolution printed image defect detection to address the abov...
Ren-Hao Zhu, Chen Zhang· Automation· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.