Visible to the public Automatic hair color de-identification

TitleAutomatic hair color de-identification
Publication TypeConference Paper
Year of Publication2015
AuthorsPrinosil, J., Krupka, A., Riha, K., Dutta, M. K., Singh, A.
Conference Name2015 International Conference on Green Computing and Internet of Things (ICGCIoT)
Keywordsautomatic hair color deidentification method, basic hair color recognition, data privacy, de-identification, Face, Facial features, Hair, Hair Color, Image color analysis, image colour analysis, image hair area segmentation, image recognition, image segmentation, multimedia content, primary biometric trait, privacy, privacy protection, pubcrawl170113, real-looking deidentified image, Segmentation, Skin, video record

A process of de-identification used for privacy protection in multimedia content should be applied not only for primary biometric traits (face, voice) but for soft biometric traits as well. This paper deals with a proposal of the automatic hair color de-identification method working with video records. The method involves image hair area segmentation, basic hair color recognition, and modification of hair color for real-looking de-identified images.

Citation Keyprinosil_automatic_2015