A mishmash of methods for mitigating the model mismatch mess

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F14%3A00225391" target="_blank" >RIV/68407700:21230/14:00225391 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1117/12.2038908" target="_blank" >http://dx.doi.org/10.1117/12.2038908</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1117/12.2038908" target="_blank" >10.1117/12.2038908</a>

Jazyk výsledku

angličtina

Název v původním jazyce

A mishmash of methods for mitigating the model mismatch mess

Popis výsledku v původním jazyce

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 andsteganalysis 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,

Název v anglickém jazyce

A mishmash of methods for mitigating the model mismatch mess

Popis výsledku anglicky

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 andsteganalysis 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,

Druh

D - Stať ve sborníku

CEP obor

JD - Využití počítačů, robotika a její aplikace

OECD FORD obor

—

Projekt

<a href="/cs/project/GPP103%2F12%2FP514" target="_blank" >GPP103/12/P514: Detekce anomalií v reálném čase a časově nestálem prostředí</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2014

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Proceedings of SPIE Media Watermarking, Security, and Forensics 2014

ISBN

978-0-8194-9945-5

ISSN

0277-786X

e-ISSN

—

Počet stran výsledku

16

Strana od-do

—

Název nakladatele

SPIE

Místo vydání

Washington

Místo konání akce

San Francisco

Datum konání akce

2. 2. 2014

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000335494300017