ETRI-Knowledge Sharing Plaform

ENGLISH

성과물

논문 검색
구분 SCI
연도 ~ 키워드

상세정보

학술대회 Adult Image Detection using Bayesian Decision Rule weighted by SVM Probability
Cited 23 time in scopus Download 4 time Share share facebook twitter linkedin kakaostory
저자
최병철, 정병호, 류재철
발행일
200911
출처
International Conference on Computer Sciences and Convergence Information Technology (ICCIT) 2009, pp.659-662
DOI
https://dx.doi.org/10.1109/ICCIT.2009.43
협약과제
09MS5400, 유해 멀티미디어 콘텐츠 분석/차단 기술개발, 정병호
초록
The SVM (support vector machine) and the SCM (skin color model) are used in detection of adult contents on images. The SVM consists of multi-class learning model and is very effective method for face detection, but complex. On the contrary, the SCM is very simple for detecting adult images using skin ratio derived from statistical characteristics of RGB color information, but less effective in close-up facial images. Hence, we propose a hybrid scheme that combines the SVM for the 1st filtering scheme using learning model (with classes of adult, benign and close-up facial images) with the SCM for the 2nd filtering scheme using skin ratio and adaptive MAP (maximum a posterior) hypothesis test based on Bayes' theorem that improves the probability of true positive detection rate of adult images. © 2009 IEEE.
KSP 제안 키워드
Bayes' theorem, Color information, Decision rules, Face detection, Facial image, Hybrid Scheme, Learning model, Maximum a Posterior(MAP), RGB color, Statistical characteristics, Support VectorMachine(SVM)