Regresar

Improve method for processing dental images with fast spatial filter and shearlet transform

Abstract:

Dental caries is one of the biggest problems in human health, in the whole world. Methods of detection are usually complex and in most cases are invasive. A non-invasive method may be using image processing. With the help of an optimized algorithm, developed in matlab, using the shearlet transformation, 2D images are processed from dental X-ray samples. The results show the efficiency of the method and the precision with which it is possible to detect the contours of the caries in all the planes that are required to analyze. For edge detection used Shearlet transform and compared with classic filters: Sobel, Prewitt, Roberts, Canny, LoG and for quality evaluation used SSIM measure, for accelerated time process for reduce noise used fast 2D Gauss filter.