Face Identifying Based On Shape Matching

Nowadays, with the rapid development of Internet and multimedia technology, image retrieval technology has become a very active area. The image retrieval technology being wildly used now is text-based image retrieval. The image retrieval system saves an image as an object, describes it with keywords or texts. The retrieval is based on matching this keywords and texts. This technology has its own defect, it ignores the content of the image and relies on the description texts which are very subjective. Contend-based image retrieval is developed to overcome this problem, in stead of using texts as index, it uses the visual content of a image as its index.The face of people is an important content of an image, and it can be used as an index of image retrieval. So, the face-matching technology becomes an important part of the content-based image retrieval .This article is focused on a face matching technology based on face shape matching. It contains three steps: Face Detection ,Facial Feature Extraction and Face Matching. Chapter 3 introduces a method of face identifying based on Angle Context algorithm.Till now, a great number of pictures has been labeled with metadata, this makes the pictures be semantic. Chapter 4 introduces a kind of relative face retrieve technology based on ontology, considering the semantic similarity.