Abstract | ||
---|---|---|
The model mismatch problem occurs in steganalysis when a binary classifier is trained on objects from one cover source and tested on another: an example of domain adaptation. It is highly realistic because a steganalyst would rarely have access to much or any training data from their opponent, and its consequences can be devastating to classifier accuracy. This paper presents an in-depth study of one particular instance of model mismatch, in a set of images from Flickr using one fixed steganography and steganalysis method, attempting to separate different effects of mismatch in feature space and find methods of mitigation where possible. We also propose new benchmarks for accuracy, which are more appropriate than mean error rates when there are multiple actors and multiple images, and consider the case of 3-valued detectors which also output `don't know'. This pilot study demonstrates that some simple feature-centering and ensemble methods can reduce the mismatch penalty considerably, but not completely remove it. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1117/12.2038908 | Proceedings of SPIE |
Keywords | Field | DocType |
steganalysis,sensors,steganography | Data mining,Steganography,Feature vector,One-class classification,Binary classification,Computer science,Mean squared error,Artificial intelligence,Steganalysis,Classifier (linguistics),Ensemble learning,Machine learning | Conference |
Volume | ISSN | Citations |
9028 | 0277-786X | 12 |
PageRank | References | Authors |
0.63 | 15 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Andrew D. Ker | 1 | 1203 | 84.75 |
Tomáš Pevný | 2 | 1043 | 45.20 |