Dental records recently used in forensic medicine to help in human identification, based mainly on radiograph images. The aim of our study is to automate this process, using image analysis and pattern recognition techniques. Postmortem radiographs, including dental radiographs, with a database of ante mortem radiographs searching, in order to get the closest match with respect to some distinct features. Contours of teeth are used as the feature for matching, since they remain more invariant over time compared to some other features of the body. The work ineludes two stages: radiograph segmentation, with pixel classification, and contour matching Probability model is used to describe the distribution of object pixels in the image. Results of retrievals on a database of over 40 images are encouraging.