期刊名称：International Journal of Engineering and Computer Science
印刷版ISSN：2319-7242
出版年度：2013
卷号：2
期号：10
页码：3075-3080
出版社：IJECS
摘要：This manuscript describes Active Appearance model for automatic Cephalometric analysis in forensic science, which isconcerned with the recognition, identification, individualization, and evaluation of physical evidence. A complete model of shape andtexture was built from a dataset of manually annotated images, and then tested with unseen images. To apply AAM Method foridentifying landmark points 140 images were collected from AAOF out of it 67 were female and 73 were male with age ranging from 15years to 45 years. By using the AAM algorithm, the mean shape is extracted and the appearance variation collected by establishing apiece-wise affine warp between each image of the training set and the mean shape.
