Title :
Towards automatic detection of child pornography
Author :
Sae-Bae, Napa ; Xiaoxi Sun ; Sencar, Husrev T. ; Memon, Nasir D.
Author_Institution :
CSE Dept., NYU, New York, NY, USA
Abstract :
This paper presents a child pornographic image detection system that identifies human skin tones in digital images, extracts features to detect explicit images and performs facial image based age classification. The novelty of the technique relies on the use of a robust and very fast skin color filter and a new set of facial features for improved identification of child faces. Tests on a dataset containing explicit images taken under different illuminations and reflecting a diversity of human skin tones, show that explicit images can be differentiated from benign images with around 90% accuracy. Similarly, tests performed on adult and child facial images yielded an accuracy of 80% in detecting child faces. Test conducted on 105 images involving semi-naked children (with no sexual context) revealed that the system has true positive rates of 83% in detecting explicit-like images and 96.5% in detecting child faces.
Keywords :
face recognition; feature extraction; image colour analysis; skin; adult facial images; automatic child pornographic image detection system; benign images; child face identification improvement; child facial images; digital images; explicit image detection; facial features; facial image based age classification; feature extraction; human skin tone identification; illuminations; seminaked children; skin color filter; true positive rates; Face detection; Feature extraction; Image color analysis; Mouth; Nose; Skin; Visualization;
Conference_Titel :
Image Processing (ICIP), 2014 IEEE International Conference on
Conference_Location :
Paris
DOI :
10.1109/ICIP.2014.7026079