基于半监督学习的眉毛图像分割方法  

Eyebrow image segmentation based on semi-supervised learning

作　　者：张晨光[1] 李玉鑑[1] 

机构地区：[1]北京工业大学计算机学院,北京100124

出　　处：《计算机工程与应用》2009年第21期139-141,共3页Computer Engineering and Applications

基　　金：国家自然科学基金(No.60775010)~~

摘　　要：眉毛图像的分割,由于受到毛发、姿势及个体差异的影响,是一个非常困难的问题。提出了一种利用半监督学习技术进行彩色眉毛图像分割的方法,首先通过手工在眉毛图像上简单画上几条线标注部分眉毛点和非眉毛点,然后利用半监督学习技术完成眉毛图像分割并从中提取纯眉毛图像,最后通过实验说明该方法具有非常好的分割效果,可用于眉毛识别的前期预处理。Eyebrow image segmentation is a very difficult problem due to the influence of hair,posture and individual eyebrow difference.This paper proposes a semi-supervised learning method to segment colorized eyebrow image.The method first labels a number of eyebrow and non-eyebrow points by manually drawing several simple lines on eyebrow image,and then uses a semisupervised learning technique to segment and extract pure eyebrow part.Experiments show that it can achieve a very good segmentation and can be used to preprocess eyebrow image before recognition.

关 键 词：半监督学习 图像分割 眉毛识别 特征提取 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]