Fast template matching algorithm for contour images based on its chain coded description applied for human face identification

摘要

A new matching algorithm for contour images described by chain coded expression is presented. In our face authentication system, the isodensity contours has been introduced to differentiate between the facial features. These isodensity contours can be transformed into chain codes. By using these coded isodensity contours, remarkable improvement in the processing performance can be expected in terms of the processing time and memory requirements.From the computer simulation performed using images of 50 people, it turned out clear that the processing time was decreased to approximately one-seventh compared to the conventional method. With respect to memory requirement, it was reduced to a quarter.