Fast algorithm for detection of reference spheres in digital panoramic radiography

Abstract In this paper, an algorithm for detection of reference spheres from digital panoramic radiographic images is presented. The proposed algorithm was tested on a database of 107 digital panoramic radiographic images which were used for dental diagnostics. Results show that the proposed method exhibits for detection of reference spheres, a sensitivity of 97.33% and specificity of 93.85%. Performance time differed between 0.55 and 2.36 s depending on image size. The aim of this work was to provide a fast ellipse detection algorithm to reduce measuring time on preoperative implant planning by lowering the computational cost.