Phoneme Recognition by Using a Sequence of Color Lip Images
Phoneme Recognition by Using a Sequence of Color Lip Images
复制标题
使用一系列彩色嘴唇图像进行音素识别
DOI：
10.1541/ieejias.119.37
复制
发表时间：
1999
期刊：
影响因子：
--
通讯作者：
S. Oe
中科院分区：
文献类型：
--
作者：
K. Terada;R. Yamanaka;S. Oe
本文提出了一种利用唇色图像序列识别音素的方法。利用设置在面部前方的彩色摄像机，对口腔图像进行序列采集。首先，在图像序列中，利用口腔运动信息检测待处理的每个目标图像。下一步，在每张图像中，通过颜色信息检测口腔区域。然后利用灰度分布的变化提取唇外线、唇内线、唇舌和唇中心线四种形状；最后，采用基于子空间的方法对音素进行识别。作为训练，子空间方法需要从上述四个形状中提取20个特征。在本文中，作者描述了实验结果，这些实验结果证明了所提出方法的有效性。
In this paper, we propose a method for recognizing phoneme by using a sequence of color lip images. The mouth images are obtained in series by using the color camera set at the front of the face. First, among the sequence of the images, each object image for processing is detected by using mouth movement information. Next step, in each image, the mouth region is detected by the color information. Then, the four shapes which are outside line of lip, the inside line of lip, the tangue and the center line of lip are extracted by using the change in distribution of gray level. Finally, the recognition of phoneme is carried out by the method based on the subspace method. As the training, the subspace method needs the 20 features extracted from the above four shapes. In this paper, the authors describe the results of the experiments which were performed to demonstrate the effectiveness of the proposed method.