Navigation

Benedikt Lorch, M. Sc.

  • Organization: Department of Computer Science
  • Working group: Chair of Computer Science 1 (IT Security Infrastructures)
  • Phone number: +49 9131 85 69918
  • Fax number: +49 9131 85 69919
  • Email: benedikt.lorch@fau.de
  • Website:
  • Address:
    Martensstr. 3
    91058 Erlangen
    Room 12.139

PhD candidate in the multimedia security group. My research interests include image forensics, computer vision, and machine learning.

Recent projects

Image forensics from chroma wrinkles

Image splicing localization where both host and donor contain the artifact, however, the artifacts in the inserted region are desynchronized with the host.

The JPEG compression format provides a rich source of forensic traces that include quantization artifacts, fingerprints of the container format, and numerical particularities of JPEG compressors. Such a diverse set of cues serves as the basis for a forensic examiner to determine origin and authenticity of an image. In this work, we present a novel artifact that can be used to fingerprint the JPEG compression library. The artifact arises from chroma subsampling in one of the most popular JPEG implementations. Due to integer rounding, every second column of the compressed chroma channel appears on average slightly brighter than its neighboring columns, which is why we call the artifact a chroma wrinkle. We theoretically derive the chroma wrinkle footprint in DCT domain, and use this footprint for detecting chroma wrinkles.
Paper | Slides

Forensic reconstruction of severely degraded license plates

Low-quality image of South Carolina license plate on car rear. The low resolution and the presence of additive noies make it nearly impossible to decipher the license number.

Forensic investigations often have to contend with extremely low-quality images that can provide critical evidence. Recent work has shown that, although not visually apparent, information can be recovered from such low-resolution and degraded images. We developed a CNN-based approach to decipher the contents of low-quality images of license plates.
Paper | Slides | GitHub

Quick links

PGP

Key ID D7994F61EECD8C75
SHA-1 Fingerprint 1781 F23E FB55 50E6 BA75 ED4E D799 4F61 EECD 8C75
Public Key ASCII-armored

Publications

2019

2017

2015