使用蜕变测试评估和增强分割模型的鲁棒性
Evaluating and Enhancing Segmentation Model Robustness with Metamorphic Testing
AI总结：
本文提出基于遗传算法的蜕变测试方法SegRMT，通过优化空间和光谱变换生成对抗样本以评估并增强图像分割模型鲁棒性，在降低模型mIoU和提升对抗训练效果上均显著优于现有基线方法。
摘要：
图像分割对于医学成像、增强现实和视频监控等应用至关重要。然而，分割模型往往缺乏鲁棒性，使其容易受到来自细微图像失真的对抗性扰动的影响。在本研究中，我们提出了SegRMT，这是一种蜕变测试方法，它利用遗传算法（GA）来优化空间和光谱变换序列，同时通过预定义的PSNR阈值保持图像保真度。使用Cityscapes数据集，我们的方法生成了有效挑战DeepLabV3分割模型的对抗性样本。我们的实验表明，SegRMT将DeepLabV3的平均交并比（mIoU）降低至6.4%，优于将mIoU降低至8.5%至21.7%之间的其他对抗性基线。此外，当用于对抗性训练时，SegRMT提升了模型性能，在专用对抗性数据集上实现了高达73%的mIoU提升，并将跨对抗性mIoU提高至53.8%，而其他方法仅为2%-10%。这些发现表明，SegRMT不仅模拟了真实的图像失真，还增强了分割模型的鲁棒性，使其成为确保安全关键应用中可靠性能的宝贵工具。
英文摘要：
Image segmentation is critical for applications such as medical imaging, augmented reality, and video surveillance. However, segmentation models often lack robustness, making them vulnerable to adversarial perturbations from subtle image distortions. In this work, we propose SegRMT, a metamorphic testing approach that leverages genetic algorithms (GA) to optimize sequences of spatial and spectral transformations while preserving image fidelity via a predefined PSNR threshold. Using the Cityscapes dataset, our method generates adversarial examples that effectively challenge the DeepLabV3 segmentation model. Our experiments show that SegRMT reduces DeepLabV3's mean Intersection over Union (mIoU) to 6.4%, outperforming other adversarial baselines that decrease mIoU to between 8.5% and 21.7%. Furthermore, when used for adversarial training, SegRMT boosts model performance, achieving mIoU improvements up to 73% on dedicated adversarial datasets and increasing cross-adversarial mIoU to 53.8%, compared to only 2%-10% for other methods. These findings demonstrate that SegRMT not only simulates realistic image distortions but also enhances the robustness of segmentation models, making it a valuable tool for ensuring reliable performance in safety-critical applications.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
