eduzhai > Applied Sciences > Engineering >

On Addressing the Impact of ISO Speed upon PRNU and Forgery Detection

  • Save

... pages left unread,continue reading

Document pages: 16 pages

Abstract: Photo Response Non-Uniformity (PRNU) has been used as a powerful devicefingerprint for image forgery detection because image forgeries can be revealedby finding the absence of the PRNU in the manipulated areas. The correlationbetween an image s noise residual with the device s reference PRNU is oftencompared with a decision threshold to check the existence of the PRNU. A PRNUcorrelation predictor is usually used to determine this decision thresholdassuming the correlation is content-dependent. However, we found that not onlythe correlation is content-dependent, but it also depends on the camerasensitivity setting. textit{Camera sensitivity}, commonly known by the name of textit{ISO speed}, is an important attribute in digital photography. In thiswork, we will show the PRNU correlation s dependency on ISO speed. Due to suchdependency, we postulate that a correlation predictor is ISO speed-specific,i.e. textit{reliable correlation predictions can only be made when acorrelation predictor is trained with images of similar ISO speeds to the imagein question}. We report the experiments we conducted to validate the postulate.It is realized that in the real-world, information about the ISO speed may notbe available in the metadata to facilitate the implementation of our postulatein the correlation prediction process. We hence propose a method calledContent-based Inference of ISO Speeds (CINFISOS) to infer the ISO speed fromthe image content.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...