Title : 
Detection and information theoretic measures for quantifying the distinguishability between multimedia operator chains
         
        
        
            Author_Institution : 
Signal Theor. & Commun. Dept., Univ. Vigo E. E. Telecomun., Vigo, Spain
         
        
        
        
        
        
            Abstract : 
Despite the large number of contributions on forensics in the recent years, there is still a lack of fundamental limits on how easily two signal processing operators (or operator chains) can be distinguished. We take a first step in this direction by proposing the use of some detection and information theoretic measures that are suitable for assessing those fundamental limits. The optimality and the links between those measures are discussed; they are used for quantifying the distinguishability between different operator chains, considering the different prior knowledge that could be available at the forensics analyst. Although this work is mainly a theoretical motivation to the use of those measures in forensic applications, its particularization to three practical scenarios is illustrated.
         
        
            Keywords : 
image forensics; multimedia computing; detection measures; forensic applications; information theoretic measures; multimedia forensics; multimedia operator chains distinguishability; signal processing operator distinguishability; Discrete cosine transforms; Forensics; Histograms; Multimedia communication; Quantization; Topology; Transform coding;
         
        
        
        
            Conference_Titel : 
Information Forensics and Security (WIFS), 2012 IEEE International Workshop on
         
        
            Conference_Location : 
Tenerife
         
        
            Print_ISBN : 
978-1-4673-2285-0
         
        
            Electronic_ISBN : 
978-1-4673-2286-7
         
        
        
            DOI : 
10.1109/WIFS.2012.6412651