Logo des Repositoriums

Auflistung nach Schlagwort "pixelized"

1 - 2 von 2
  • Textdokument
    Recovering information from pixelized credentials
    (GI SICHERHEIT 2022, 2022) Garske, Viktor; Noack, Andreas
    Pixelation is a common technique to redact sensitive information like credentials in images. In this paper, we propose a system that is able to recover information from pixelized text. Our contribution consists of a neural network as well as a generic pipeline that generates a realistic training dataset considering flexible specifications including wordlists, fonts, font sizes and letter spacings. The contributed neural network is a composition of a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN) using Long short-term memory (LSTM) and a Connectionist Temporal Classification (CTC) layer to decode sequences of characters. With our approach, we achieve a Label Error Rate (LER) under 50% when taking pixelation block sizes of up to 8 × 8 pixels on a 22pt font into account. Thereby, our results indicate that pixelation of sensitive data does not satisfy common privacy standards.
  • Textdokument
    Revisiting the privacy of censored credentials
    (INFORMATIK 2022, 2022) Garske,Viktor; Noack,Andreas
    On the internet, you find numerous images like screenshots where secret parts are hidden with irreversible redaction techniques like pixelation or blurring. In this paper, we propose a system that recovers information from redacted text in raster graphics using a composition of a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN) using Long short-term memory (LSTM) and a Connectionist Temporal Classification (CTC) layer to output the most probable character sequence. We furthermore show that our model operates in an automated pipeline, performs on blurred images without modification and is even able to compensate JPEG quality loss. Finally, our test results indicate that a generic neural network can be trained successfully to assist the recovery of pixelized or blurred information on screenshots or high-quality photos.