Morphology-Aware Sample Assignment: Overcoming IoU Insensitivity for Surface Defect Detection
Quick Answer
The study introduces morphology-aware sample assignment to enhance surface defect detection in visual models like YOLOv9, addressing IoU insensitivity.
Quick Take
The study introduces morphology-aware sample assignment to enhance surface defect detection in visual models like YOLOv9, addressing IoU insensitivity. By integrating morphological metrics, the method improves positive sample precision without additional inference costs, demonstrating performance gains on NEUDET and GC10-DET datasets.
Key Points
- Introduces morphological similarity metrics to refine positive sample assignment.
- Achieves performance gains on NEUDET and GC10-DET datasets using YOLOv9.
- No additional inference overhead ensures deployment efficiency in industrial settings.
- Addresses IoU insensitivity by reshaping response distribution for better matching.
- Enhances precision of positive sample selection significantly.
Paper Resources
Article Content
From source RSS / original summaryarXiv:2606. 13723v1 Announce Type: new Abstract: Intersection-over-Union (IoU), as a pivotal metric for evaluating the spatial alignment between candidate proposals and ground-truth annotations, directly determines the quality of positive sample sets and the training efficacy of visual detection models. Through theoretical modeling and analysis, we uncover a non-sensitive region on the IoU response curve, within which samples yield nearly identical IoU scores despite distinct geometric overlaps.
To overcome this limitation, we introduce a set of morphological similarity metrics covering area, shape, and aspect ratio, to refine the positive sample assignment process, thereby ensuring more discriminative and reliable matching. A supplementary matching score is derived via mean-based aggregation of these multidimensional similarities, compensating for the intrinsic limitation of IoU in representing structural correspondence.
Theoretically, incorporating morphological similarity reshapes the response distribution of the matching function, yielding both effective directional gradients and polygon-like iso-response contours, which tightly confine high-response regions around each ground-truth instance and substantially enhance the precision of positive sample selection. Experiments based on the YOLOv9 framework demonstrate consistent performance gains on both NEUDET and GC10- DET datasets.
Notably, the proposed approach is fully plug-and-play and incurs zero additional inference overhead, thereby ensuring deployment efficiency for industrial visual inspection.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →LLM-Guided ANN Index Optimization for Human-Object Interaction Retrieval
A phase-aware LLM agent optimizes human-object interaction retrieval, outperforming Optuna TPE by 33.3% and VDTuner by 34.2% on the HICO-DET benchmark. This method enhances throughput by 15.3x over UniIR and demonstrates strong transferability across vector database management systems.