Google's Cloud Vision API Is Not Robust To Noise

Abstract

Google has recently introduced the Cloud Vision API for image analysis. According to the demonstration website, the API "quickly classifies images into thousands of categories, detects individual objects and faces within images, and finds and reads printed words contained within images." It can be also used to "detect different types of inappropriate content from adult to violent content." In this paper, we evaluate the robustness of Google Cloud Vision API to input perturbation. In particular, we show that by adding sufficient noise to the image, the API generates completely different outputs for the noisy image, while a human observer would perceive its original content. We show that the attack is consistently successful, by performing extensive experiments on different image types, including natural images, images containing faces and images with texts. For instance, using images from ImageNet dataset, we found that adding an average of 14.25% impulse noise is enough to deceive the API. Our findings indicate the vulnerability of the API in adversarial environments. For example, an adversary can bypass an image filtering system by adding noise to inappropriate images. We then show that when a noise filter is applied on input images, the API generates mostly the same outputs for restored images as for original images. This observation suggests that cloud vision API can readily benefit from noise filtering, without the need for updating image analysis algorithms.

References in corpus (5)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Understanding Adversarial Training: Increasing Local Stability of Neural Nets through Robust Optimization
• Practical Black-Box Attacks against Machine Learning
• Blocking Transferability of Adversarial Examples in Black-Box Learning Systems
• Deceiving Google's Cloud Video Intelligence API Built for Summarizing Videos

Cited by in corpus (4)

• Large Margin Deep Networks for Classification
• Transferability of Adversarial Examples to Attack Cloud-based Image Classifier Service
• Cloud-based Image Classification Service Is Not Robust To Simple Transformations: A Forgotten Battlefield
• Sparsifying and Down-scaling Networks to Increase Robustness to Distortions