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.
Abstract (translated)
公用事业公司越来越多地依赖无人机图像进行事后的和常规的检查，但是由于缺陷实例稀少且检查数据集往往有限或专有，训练准确的缺陷分类器仍然具有挑战性。我们通过使用现成的多模态大型语言模型（MLLM）作为无需训练的图像生成器来解决这一数据稀缺问题，该方法可以从视觉参考和文本提示中合成缺陷图像。我们的流程通过双参考条件增加多样性、通过轻量级的人工验证和提示改进提高标签准确性，并利用基于嵌入的选择规则过滤出最终的合成池，此选择规则基于计算真实训练集各类质心的距离。 我们在一个具有现实低数据量设定（104张真实的训练图像；152张验证图像；308张测试图像）的公共陶瓷绝缘子缺陷类型分类（壳体 vs. 釉面）的数据集上进行了评估。将10%的真实训练集与基于嵌入选择的合成图像相结合，使测试F1分数（精度和召回率的调和平均数）从0.615提高到了0.739（相对提高了20%），相当于估计出4-5倍的数据效率增益，并且这种改进在使用更强的基础模型和支持冻结特征线性探测基准时依然持续存在。 这些结果表明，在收集更多真实缺陷变得缓慢或不可行的情况下，有一条实用、门槛较低的路径可以提高缺陷识别能力。
URL
https://arxiv.org/abs/2603.08069
