Dynamic time warping (DTW) algorithm towards accurate radiographic images diagnostic system

---

Somaya Adwan

Arab Open University, KSA

: J Diagn Tech Biomed Anal

Abstract

---

T ypical X-ray images produced using conventional screen-film procedure has a limited field of view and failed to achieve the entire long bone visualization on a single image. To produce images with whole spinal parts, digitized images from the films contain portions of the spinal parts are merged using image stitching methods. Panorama image creation is a process of over-laying a set of images taken at different view-points and different time into one coordinate system to generate a wider viewing panoramic image. The most important step in panorama image is image stitching whose components are image registration and image blending. In image registration, portions of consecutive images are modeled to find a transformation position which aligns the images. In this paper, a novel Image Stitching method, which utilizes dynamic time warping algorithm that provides a measure of similarity and matching points between a numbers of radiographic X-ray images is developed to provide accurate spinal deformities diagnosis. The effectiveness of the proposed method is demonstrated in the experiments. As a result, DTW is able to detect the similar points in stitched images. It is found that the proposed technique achieves high detection accuracy with less processing time.

Biography

---

E-Mail:

somi@siswa.um.edu.my