An image classification system defends against physically realizable attacks. A training dataset of input images is retrieved and an adversarial image is generated based on one of the input images that is selected. The adversarial image is created by occluding a portion of the selected image by superimposing a predetermined shape (e.g., a rectangle) containing noise on the selected image. A defense against occlusion attacks (DOA) classifier is trained using the training dataset and the adversarial image. The DOA classifier is utilized to classify captured images of items (e.g., street signs) that may have been attacked (e.g., sticker placement, vandalism).


    Access

    Download


    Export, share and cite



    Title :

    SYSTEMS AND METHODS FOR DEFENDING AGAINST PHYSICAL ATTACKS ON IMAGE CLASSIFICATION


    Contributors:
    VOROBEYCHIK YEVGENIY (author) / WU TONG (author) / TONG LIANG (author)

    Publication date :

    2021-09-30


    Type of media :

    Patent


    Type of material :

    Electronic Resource


    Language :

    English


    Classification :

    IPC:    B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion / G06N COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS , Rechnersysteme, basierend auf spezifischen Rechenmodellen / G06V



    Systems and methods for defending against physical attacks on image classification

    VOROBEYCHIK YEVGENIY / WU TONG / TONG LIANG | European Patent Office | 2025

    Free access

    Defending Commercial Airliners against Missile Attacks

    Hepp, B. / Ravat, C. / North Atlantic Treaty Organization | British Library Conference Proceedings | 2005


    Defending Transportation Networks against Random and Targeted Attacks

    Lou, Yingyan / Zhang, Lihui | Transportation Research Record | 2011



    METHOD, APPARATUS, DEVICE, AND STORAGE MEDIUM FOR DEFENDING AGAINST ATTACKS

    CUI ZHIYANG | European Patent Office | 2021

    Free access