论文标题
使用两因素扰动对深度学习分类器的鲁棒性进行基准测试
Benchmarking Robustness of Deep Learning Classifiers Using Two-Factor Perturbation
论文作者
论文摘要
深度学习（DL）分类器的准确性通常是不稳定的，因为在对抗图像，不完美的图像或扰动图像上重新测试时，它们可能会发生重大变化。本文增加了基准在有缺陷图像上基准DL分类器的鲁棒性的基本工作。为了衡量强大的DL分类器，先前的研究报告了单因素腐败。我们创建了全面的69个基准图像集，包括一个清洁集，具有单个因子扰动的集合以及具有两因素扰动条件的集合。最先进的两因素扰动包括（a）两种序列中应用的两个数字扰动（盐和胡椒噪声和高斯噪声），以及（b）在两个序列中应用的一个数字扰动（盐和胡椒噪声）和几何扰动（盐和胡椒噪声）和几何扰动（旋转）。先前评估DL分类器的研究经常使用TOP-1/TOP-5精度。我们创新了一个新的二维统计矩阵，以评估DL分类器的鲁棒性。此外，我们引入了一种新的可视化工具，包括最低准确性，最大精度，平均精度和变化系数（CV），用于基准DL分类器的鲁棒性。与单因素损坏相比，我们首先报告说，使用两因素扰动图像可以提高DL分类器的鲁棒性和准确性。所有源代码和相关图像集均在网站上共享http://cslinux.semo.edu/david/data，以支持未来的学术研究和行业项目。
Accuracies of deep learning (DL) classifiers are often unstable in that they may change significantly when retested on adversarial images, imperfect images, or perturbed images. This paper adds to the fundamental body of work on benchmarking the robustness of DL classifiers on defective images. To measure robust DL classifiers, previous research reported on single-factor corruption. We created comprehensive 69 benchmarking image sets, including a clean set, sets with single factor perturbations, and sets with two-factor perturbation conditions. The state-of-the-art two-factor perturbation includes (a) two digital perturbations (salt & pepper noise and Gaussian noise) applied in both sequences, and (b) one digital perturbation (salt & pepper noise) and a geometric perturbation (rotation) applied in both sequences. Previous research evaluating DL classifiers has often used top-1/top-5 accuracy. We innovate a new two-dimensional, statistical matrix to evaluating robustness of DL classifiers. Also, we introduce a new visualization tool, including minimum accuracy, maximum accuracy, mean accuracies, and coefficient of variation (CV), for benchmarking robustness of DL classifiers. Comparing with single factor corruption, we first report that using two-factor perturbed images improves both robustness and accuracy of DL classifiers. All source codes and related image sets are shared on the Website at http://cslinux.semo.edu/david/data to support future academic research and industry projects.
