This paper proposes a novel approach for automating the analysis of identifying the person based on their ante mortem and postmortem reports. This approach involves three techniques (i.e.) morphological contour detector, Gaussian filtering and an existing semi-automatic contour extraction method. Forensic dentistry involves the identification of people based on their dental records, mainly available as radiograph images. Our goal is to automate this process using image processing and pattern recognition techniques. Given a postmortem radiograph, we search a database of antemortem radiographs in order to retrieve the closest match with respect to some salient features. In this paper, we use the contours of the teeth as the feature for matching. The algorithm completes the task in three steps: radiograph segmentation, pixel classification and contour matching. In this paper a hit rate of 0.7 is achieved by the Morphological contour detectors which are comparable with the other two methods.

No abstract

No abstract

This paper proposes an automated target recognition algorithm using Support Vector Machine (SVM) to extract landmark points for craniofacial features in cephalometry radiograph. The features are extracted by subjecting the radiograph to the Projected Principle Edge Distribution (PPED) algorithm. Edge flags are accumulated in every four columns and spatial distribution of edge flags are represented by a histogram. The resultants are the front end of support vector machine. Vectors, which possess land marks, are separated from all other vectors. The centroid points, automatically determined from PPED vectors, are the location of landmarks. The landmark points which are serving as a guide for construction and measurement of planes, are used to evaluate the dento-facial relationship, study of growth and development, and also for treatment planning.

No abstract