Publication Type
Conference Paper
Book Title
Proceedings of the Workshop on Artificial Intelligence Safety 2019, co-located with the 28th International Joint Conference on Artificial Intelligence (IJCAI-19) [http://ceur-ws.org]
Publication Date
Page Number
25
Volume
Vol. 2419
Conference Name
AISafety 2019 (held in conjunction with IJCAI 2019 - International Joint Conference on Artificial Intelligence)
Conference Location
Macao, China
Conference Sponsor
Sony (Main Sponsor)
Conference Date
-
Abstract
Deep learning systems are increasingly being adopted for safety critical tasks such as autonomous driving. These systems can be exposed to adverse weather conditions such as fog, rain and snow. Vulnerability of deep learning systems to synthetic adversarial attacks has been extensively studied and demonstrated, but the impact of natural weather conditions on these systems has not been studied in detail. In this paper, we study the effects of fog on classification accuracy of the popular Inception deep learning model. We use stereo images from the Cityscapes dataset and computer graphics techniques to mimic realistic naturally occurring fog. We show that the Inception deep learning model is vulnerable to the addition of fog in images.