Human skin detection through correlation rules between the YCb and YCr subspaces based on dynamic color clustering

详细信息    查看全文

• 作者：Nadia Brancati ; ^{nadia.brancati@cnr.it} ; ^{nadia.brancati@na.icar.cnr.it} ; Giuseppe De Pietro ^{giuseppe.depietro@cnr.it} ; Maria Frucci ^{maria.frucci@cnr.it} ; Luigi Gallo ^{luigi.gallo@cnr.it}
• 关键词：Skin detection ; YCbCr color space ; Pixel based method ; Pixel classification ; Dynamic color clustering ; Correlation rules
• 刊名：Computer Vision and Image Understanding
• 出版年：2017
• 出版时间：February 2017
• 年：2017
• 卷：155
• 期：Complete
• 页码：33-42
• 全文大小：3834 K
• 卷排序：155

文摘

Novel skin detection method based on correlation rules that evaluate the combinations of chrominance values to identify the skin pixels in the YCb and YCr subspaces Definition of a fitting model of skin-candidate pixels for the generation of image-specific trapezia in the YCb and YCr subspaces Comparisons with six well-known rule-based methods in literature show that the proposed method outperforms the others in terms of F-measure, Dprs and Precision