A Segmentation-driven Editing Method for Bolt Defect Augmentation and Detection

2025-08-14

AI Summary

To address the challenges of scarce and severely imbalanced bolt defect images in transmission line inspection—leading to suboptimal detection performance—this paper proposes a segmentation-driven defect editing augmentation framework. The method first leverages high-precision attribute-aware segmentation, enhanced by a novel Channel-wise Feature Alignment (CFA) module and a Multi-Aspect Mask Decoder (MAMD) to improve segmentation robustness. It then integrates MOD-LaMa for fine-grained, photorealistic defect synthesis and introduces a scene-adaptive Editing-to-Realistic Adaptation (ERA) strategy to seamlessly embed edited defects into original inspection backgrounds. Evaluated on a self-constructed multi-source bolt dataset, the framework significantly outperforms existing editing-based approaches, boosting mAP of mainstream detectors—including YOLOv8 and Faster R-CNN—by 6.2–9.7% on average. This work establishes a transferable technical pathway for few-shot industrial defect detection.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Feature Construction/Reformulation
• Search and Optimization: Learning to Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work

Abstract

Bolt defect detection is critical to ensure the safety of transmission lines. However, the scarcity of defect images and imbalanced data distributions significantly limit detection performance. To address this problem, we propose a segmentationdriven bolt defect editing method (SBDE) to augment the dataset. First, a bolt attribute segmentation model (Bolt-SAM) is proposed, which enhances the segmentation of complex bolt attributes through the CLAHE-FFT Adapter (CFA) and Multipart- Aware Mask Decoder (MAMD), generating high-quality masks for subsequent editing tasks. Second, a mask optimization module (MOD) is designed and integrated with the image inpainting model (LaMa) to construct the bolt defect attribute editing model (MOD-LaMa), which converts normal bolts into defective ones through attribute editing. Finally, an editing recovery augmentation (ERA) strategy is proposed to recover and put the edited defect bolts back into the original inspection scenes and expand the defect detection dataset. We constructed multiple bolt datasets and conducted extensive experiments. Experimental results demonstrate that the bolt defect images generated by SBDE significantly outperform state-of-the-art image editing models, and effectively improve the performance of bolt defect detection, which fully verifies the effectiveness and application potential of the proposed method. The code of the project is available at https://github.com/Jay-xyj/SBDE.

Problem

• Augmenting bolt defect images for better detection
• Improving segmentation of complex bolt attributes
• Converting normal bolts to defective ones via editing

Innovation

• Segmentation-driven bolt defect editing method
• CLAHE-FFT Adapter and Multipart-Aware Mask Decoder
• Editing recovery augmentation strategy