HUMAN SKIN DETECTION USING GAUSSIAN MIXTURE MODEL

被引：0

作者：

Oancea, Romana ^{[1
]}

Demeter, Stefan ^{[1
]}

Kifor, Stefania ^{[2
]}

机构：

[1] Nicolae Balcescu Land Forces Acad, Sibiu, Romania

[2] Lucian Blaga Univ Sibiu, Sibiu, Romania

来源：

15TH INTERNATIONAL CONFERENCE THE KNOWLEDGE-BASED ORGANIZATION: APPLIED TECHNICAL SCIENCES AND ADVANCED MILITARY TECHNOLOGIES, CONFERENCE PROCEEDINGS 6 | 2009年 / 6卷

关键词：

skin detection; Gaussian mixture; color space; EM algorithm;

D O I：

暂无

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

For the tracking and recognition of the human faces in static images, one can chose, during the preprocessing stage, to delimitate the skin areas. Then, these areas, obtained during the preprocessing stage, will be analyzed by techniques like template matching, feature detection or neural network in order to select only the human faces. In the present study we tried to create a skin model using Gaussian mixtures, a parametric method often encountered in the specialized literature.

引用

页码：113 / 118

页数：6

共 7 条

[1]

CAETANO TS, 2003, PATTERN RECOGNITION

[2]

GOKALP D, LEARNING SKIN PIXELS

[3]

JONES MJ, 1999, STAT COLOR MODELS AP, V1, P274

[4] MIXTURE DENSITIES, MAXIMUM-LIKELIHOOD AND THE EM ALGORITHM [J].

REDNER, RA ;

WALKER, HF .

SIAM REVIEW, 1984, 26 (02) :195-237

[5]

YANG M, 1999, P SPIE, V3656

[6]

YANG MH, 1998, P IEE INT C IM PROC

[7]

Zarit B. D., 1999, Proceedings International Workshop on Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems. In Conjunction with ICCV'99 (Cat. No.PR00378), P58, DOI 10.1109/RATFG.1999.799224

← 1 →