Multi-orientation local ternary pattern-based feature extraction for forensic dentistry

Abstract Accurate and automated identification of the deceased victims with dental radiographs plays a significant role in forensic dentistry. The image processing techniques such as segmentation feature extraction play crucial retrieval accordance matching image. raw undergoes segmentation, distance-based retrieval. ultimate goal proposed work is quality enhancement by providing advanced techniques, extraction, techniques. In this paper, multi-orientation local ternary pattern-based for extraction. grey level difference method (GLDM) adopted to extract texture shape features that are considered better results. done computation similarity score using distances Manhattan, Euclidean, vector cosine angle, histogram intersection distance obtain optimal match from database. manually picked dataset 200 images performance analysis. By extracting both features, approach achieved maximum accuracy, precision, recall, F-measure, sensitivity, specificity lower false-positive negative values.