Recently, Deep Neural Networks (DNNs) have achieved remarkable performances in many applications, while several studies have enhanced their vulnerabilities to malicious attacks. In this paper, we emulate the effects of natural weather conditions to introduce plausible perturbations that mislead the DNNs. By observing the effects of such atmospheric perturbations on the camera lenses, we model the patterns to create different masks that fake the effects of rain, snow, and hail. Even though the perturbations introduced by our attacks are visible, their presence remains unnoticed due to their association with natural events, which can be especially catastrophic for fully-autonomous and unmanned vehicles. We test our proposed fakeWeather attacks on multiple Convolutional Neural Network and Capsule Network models, and report noticeable accuracy drops in the presence of such adversarial perturbations. Our work introduces a new security threat for DNNs, which is especially severe for safety-critical applications and autonomous systems.


    Zugriff

    Download


    Exportieren, teilen und zitieren



    Titel :

    fakeWeather: Adversarial Attacks for Deep Neural Networks Emulating Weather Conditions on the Camera Lens of Autonomous Systems



    Erscheinungsdatum :

    2022-01-01



    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Klassifikation :

    DDC:    629



    PHYSICAL ADVERSARIAL ATTACKS ON DEEP NEURAL NETWORKS FOR TRAFFIC SIGN RECOGNITION: A FEASIBILITY STUDY

    Woitschek, Fabian / Schneider, Georg | British Library Conference Proceedings | 2021




    Traffic Sign Recognition Robustness in Autonomous Vehicles Under Physical Adversarial Attacks

    Apostolidis, Kyriakos D. / Gkouvrikos, Emmanouil V. / Vrochidou, Eleni et al. | Springer Verlag | 2023


    Detection of Physical Adversarial Attacks on Traffic Signs for Autonomous Vehicles

    Villarini, Barbara / Radoglou-Grammatikis, Panagiotis / Lagkas, Thomas et al. | IEEE | 2023