关键词: 半监督学习, 图像分割, 眉毛识别, 特征提取

Abstract: 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 semi-supervised 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.

Key words: semi-supervised, image segmentation, eyebrow recognition, feature extraction