This paper investigates strategies that defend against adversarial-example
attacks on image-classification systems by transforming the inputs before
feeding them to the system. Specifically, we study applying image
transformations such as bit-depth reduction, JPEG compression, total variance
minimization, and image quilting before feeding the image to a convolutional
network classifier. Our experiments on ImageNet show that total variance
minimization and image quilting are very effective defenses in practice, in
particular, when the network is trained on transformed images. The strength of
those defenses lies in their non-differentiable nature and their inherent
randomness, which makes it difficult for an adversary to circumvent the
defenses. Our best defense eliminates 60% of strong gray-box and 90% of strong
black-box attacks by a variety of major attack methods

探讨了采用图像转换策略，如降低位深、JPEG 压缩、总变量最小化和图像镶嵌等方法，以防御对图像分类系统的对抗性攻击。在 ImageNet 上的实验表明，总变量最小化和图像拼接是非常有效的防御方法，在网络对转换后的图片进行训练时表现尤为突出。最好的防御方法可以消除各种主要攻击方法的 60% 的强度灰盒和 90% 的强度黑盒攻击。