Description

It is an novel method to detect the human by the use their face. These are achieved by extracting the features of the face. Initially the databases are loaded. From databases cropped images are segmented by the background elimination. After extraction, we extract the feature as height, width of the human and six ankles of the human (head to neck(1),neck to torso(1),torso to knee(2),knee to toe(2)). The RVM will be trained by the dataset feature. The trained feature will classify by the RVM. Finally it will recognize the person.