Quality factor estimation of JPEG images using a statistical model

Florent Retraint (1) , Cathel Zitzmann (2)

1 LM2S - Laboratoire Modélisation et Sûreté des Systèmes
2 EPF - Ecole Polytechnique Féminine

This paper proposes a new approach to quality factor identification for a JPEG images. This approach is based on a spatial domain model of the variance of 8*8 blocks of JPEG images. The variance model follows a Gamma distribution, which is characterized by two parameters. These parameters are considered as a unique fingerprint for determining the value of the quality factor. To demonstrate the benefits of this approach, a likelihood ratio (LR) is established by exploiting the variance distribution. This ratio enables the quality factor to be obtained. Experiments on simulated images and large image databases with different image sizes and quality factors highlight the high accuracy of the proposed method.

Keywords: Digital forensics; JPEG compression history; Variance distribution; Quality factor estimation; Statistical image modeling

Domains: Signal and Image processing

Florent Retraint, Cathel Zitzmann. Quality factor estimation of JPEG images using a statistical model. Digital Signal Processing, 2020, 103, pp.102759. ⟨10.1016/j.dsp.2020.102759⟩. ⟨hal-02571650⟩