Research on lip detection based on Opencv

Lip-reading technology play an important role to improve the recognition rate of the speech recognition systems on the noise environment and to help the disabled in hearing communicate with others. Because the shape of lip has obvious changes when somebody is pronouncing, it is difficult to directly detect the lip region; on the contrary, the nose shape can not be able to have the obvious changes, and the face detection in opencv software has achieved considerable effect on detection face, so the paper has presented a method of detection lip; this method adopts the relative position of lip against to face and nose to detect the lip region. The experiments show that this method can quickly and efficiently extract the lip region, and improve the lip segmentation accurately.