Several kinds of word recognition techniques have been investigated and some of them are put to practical use. According to these techniques, relatively high recognition rates can be achieved in calm circumstances. But in noisy circumstances, it is very hard to accomplish high recognition rate. Considering these points, liP reading systems are proposed. Such lip reading systems recognize the lip shape movements and are not affected by surrounding noise. For development of these lip reading systems, it is indispensable to extract lip areas from face images correctly. In this paper, the novel distribution function to distinguish hp area from face images is proposed. The technique to control parameters of new distribution function by using fuzzy theory is also proposed. Experimental results are included to confirm the effectiveness of proposed technique.
展开▼
