A new knowledge-based face image indexing system through the Internet

Image indexing is one of the most popular research areas of computer vision, such as for digital libraries, multimedia systems and TV program systems. This paper focuses on the face image indexing applications. The authors proposed a new scheme for face image indexing by using a knowledge-based method that artists draw human portraits. The basic idea is to analyze artists how to draw a portrait based on human face proportions related knowledge. And then, we use these rules to process face image indexing. This process provides a new alternative way for indexing human face image in terms of the knowledge-based method. Our system is composed of two main steps. The first step is facial features extracting. And then, it goes to human face image indexing. There are two advantages of this system. First, the system provides a simple synthesis procedure. Secondly, the paper proposes an easy method which is important for reducing waiting time of delaying cost of Internet server.