Blind Image Source Camera Identification

EasyChair Preprint 5713

Abstract

The creation and control of digital images are simplified by computerized preparing apparatuses that are effective and generally accessible. As an outcome, we can at this point don't take the legitimacy of the pictures. This is particularly obvious with regards to legitimate photographic proof. With an ever-increasing computerized crime percentage, the opportunity has already come and gone. This project depicts how digital legal procedures for source examination and identification empower measurable experts to plan a picture under an inquiry to its source camera, in a visually impaired way, with no apriori data about the storage and preparing. Even though Photo Response Non-Uniformity (PRNU) has acquired an incredible interest in image forensics the extraction of camera fingerprint with minimum seen content is still a challenging problem. In our proposed method fingerprint is extracted using a weighted combination of two types of de-noising filters, weighted nuclear norm minimization (WNNM) based filter and wavelet filter. A set of images captured by different mobile cameras at different locations are used as a dataset. Each image is distinct and contains different seen details having varying exposures. After the fingerprint extraction process, the fingerprints are clustered using spectral clustering based on their origin.

Keyphrases: Fingerprint, Photoresponse non-uniformity (PRNU), Weighted nuclear norm minimization (WNNM)

@booklet{EasyChair:5713,
  author    = {K Shimji and M Baburaj},
  title     = {Blind Image Source Camera Identification},
  howpublished = {EasyChair Preprint 5713},
  year      = {EasyChair, 2021}}