Logo des Repositoriums

Auflistung nach Autor:in "Dittmar, Christian"

1 - 4 von 4
  • Konferenzbeitrag
    Combining ENF phase discontinuity checking and temporal pattern matching for audio tampering detection
    (INFORMATIK 2013 – Informatik angepasst an Mensch, Organisation und Umwelt, 2013) Mann, Sebastian; Cuccovillo, Luca; Aichroth, Patrick; Dittmar, Christian
    In this paper, we present an improved approach for audio tampering detection and localization based on the Electrical Network Frequency (ENF) analysis, combining analysis of the ENF phase, and ENF temporal pattern matching: The proposed algorithm uses phase discontinuity checking to detect regions that might have been tampered, which are then matched against an ENF reference database to validate order and duration of the detected regions. Using this approach, the false-positive rate can be reduced from ≈ approx 55% using phase analysis to ≈ approx 10% using the combined approach, thus improving overall reliability of the tampering detection approach.
  • Konferenzbeitrag
    Estimating MP3PRO encoder parameters from decoded audio
    (INFORMATIK 2013 – Informatik angepasst an Mensch, Organisation und Umwelt, 2013) Bießmann, Paul; Gärtner, Daniel; Dittmar, Christian; Aichroth, Patrick; Schnabel, Michael; Schuller, Gerald; Geiger, Ralf
    We present an approach to estimate encoder parameters from previously MP3PRO-compressed, then decompressed audio material. The algorithm has been designed to identify the presence of spectral band replication (SBR), and for bit-rate detection. Furthermore, MP3 compression parameters like frame offset and block type are detected. As evaluation results show, the approach is able to identify SBR with a high accuracy, while the compression bit-rate detection is prone to errors, especially for higher bit-rates.
  • Konferenzbeitrag
    Evaluation of an Image and Music Indexing Prototype
    (Workshop Audiovisuelle Medien WAM 2009, 2009) Dunker, Peter; Paduschek, Ronny; Dittmar, Christian; Nowak, Stefanie; Gruhne, Matthias
    This paper describes a technical solution for automated semantic indexing of music and images for a media archive environment. The indexing is based on a multi-modal low-level feature extraction and semantic high-level feature classification such as mood, genre, daytime or visual scene types. The classification on both, the audio and the visual information is based on a generic machine learning core architecture. A combination and cleansing process validates for improving the classification results. This paper presents the technical realization of a prototype and its corresponding evaluation. Finally, the practical relevance of this technology results, based on the findings of the evaluation is discussed.
  • Textdokument
    Workshop WS01 „Musik trifft Informatik“
    (INFORMATIK 2017, 2017) Müller, Meinard; Dittmar, Christian