Por favor, use este identificador para citar o enlazar este ítem: http://cio.repositorioinstitucional.mx/jspui/handle/1002/201
SEGMENTATION OF MOLARS IN NOISY PANTOMOGRAMS USING DIGITAL IMAGE PROCESSING TECHNIQUES
José Luis Reynoso Rodríguez
Francisco Javier Cuevas-de-la-Rosa
Acceso Abierto
Atribución-NoComercial-SinDerivadas
DIGITAL IMAGE PROCESSING, MATLAB, SEGMENTATION, TEETH, XRAYS, PANTOMOGRAMS.
"Digital image processing techniques are useful in order to automatically detect certain patterns in images. In the case of dental x-rays, there is no work done to automate the detection of particular teeth and measurement of its properties. In this work, two different approaches are used to isolate molars from the rest of the teeth in a panoramic x-ray. In this thesis, digital image processing techniques were used in order to isolate and segment individual molars in pantomograms; in particular, second and third molars. This is done using two different approaches pre-processing the image. The general area of the tooth is cropped beforehand manually in order to speed-up the digital image processing techniques. The orientation of the segmented molars is also determined using image moments. In both implemented methods, the pantomograms are preprocessed by applying a Gaussian filter to smooth some edges of the image. The smoothed image is then clustered into superpixels, and the mean color of each cluster is computed. Each pixel is then assigned its cluster’s mean color. This helps to achieve color uniformity and reduces noise while preserving important image information such as edges."
2017-05
Trabajo de grado, maestría
Inglés
León, Guanajuato
Público en general
Reynoso Rodríguez, (2017). "Segmentation of molars in noisy pantomograms using digital image processing techniques". Tesis de Maestría en Optomecatrónica. Centro de Investigaciones en Óptica, .A.C. León, Guanajuato. 81 pp.
TRATAMIENTO DIGITAL. IMÁGENES
Aparece en las colecciones: MAESTRIA EN OPTOMECATRÓNICA

Cargar archivos:


Fichero Descripción Tamaño Formato  
17048.pdfTexto completo / Full text2.69 MBAdobe PDFVisualizar/Abrir