Synthetic Defect Image Generation for Power Line Insulator Inspection Using Multimodal Large Language Models

TL;DR

This paper presents a method using multimodal large language models to generate synthetic defect images for power line insulator inspection, significantly enhancing classification performance in low-data scenarios.

Contribution

It introduces a novel pipeline leveraging off-the-shelf multimodal models for synthetic image generation, filtering, and augmentation to improve defect classification with limited real data.

Findings

Synthetic images boost F1 score from 0.615 to 0.739.

Data-efficiency improved by 4-5 times with synthetic augmentation.

Method remains effective with stronger models and linear probes.

Abstract

Utility companies increasingly rely on drone imagery for post-event and routine inspection, but training accurate defect-type classifiers remains difficult because defect examples are rare and inspection datasets are often limited or proprietary. We address this data-scarcity setting by using an off-the-shelf multimodal large language model (MLLM) as a training-free image generator to synthesize defect images from visual references and text prompts. Our pipeline increases diversity via dual-reference conditioning, improves label fidelity with lightweight human verification and prompt refinement, and filters the resulting synthetic pool using an embedding-based selection rule based on distances to class centroids computed from the real training split. We evaluate on ceramic insulator defect-type classification (shell vs. glaze) using a public dataset with a realistic low training-data regime (104 real training images; 152 validation; 308 test). Augmenting the 10% real training set with embedding-selected synthetic images improves test F1 score (harmonic mean of precision and recall) from 0.615 to 0.739 (20% relative), corresponding to an estimated 4--5x data-efficiency gain, and the gains persist with stronger backbone models and frozen-feature linear-probe baselines. These results suggest a practical, low-barrier path for improving defect recognition when collecting additional real defects is slow or infeasible.