Crest Lines Extraction in Volume 3d Medical Images : a Multi-scale Approach
نویسندگان
چکیده
Recently, we have shown that the diierential properties of the surfaces represented by 3D volumic images can be recovered using their partial derivatives. For instance, the crest lines can be characterized by the rst, second and third partial derivatives of the grey level function I(x; y; z). In this paper, we show that : the computation of the partial derivatives of an image can be improved using recursive lters which approximate the Gaussian lter, a multi-scale approach solves many of the instability problems arising from the computation of the partial derivatives, we illustrate the previous point for the crest line extraction (a crest point is a zero-crossing of the derivative of the maximum curvature along the maximum curvature direction). We present experimental results of crest point extraction on synthetic and 3-D medical data. Extraction de lignes de cr^ etes dans des images volumiques 3D m edicales : une approche multi-echelle R esum e : R ecemment, Nous avons montr e que les propri et es dii erentielles des surfaces peuvent ^ etre calcul ees a partir des d eriv ees partielles des images volumiques 3D. Par exemple, les lignes de cr^ ete peuvent ^ etre caract eris ees a l'aide des premi eres, deuxi emes et troisi emes d eriv ees partielles de la fonction des niveaux de gris I(x; y; z). Dans cet article nous montrons que : le calcul des d eriv ees parielles d'une image peut ^ etre am elior e en utilisant des ltres r ecursifs approximant le ltre gaussien et ses d eriv ees, une approche multi-echelle r esoud un bon nombre des probl emes d'instabilit e li es au calcul des d eriv ees partielles, nous illustrons le pr ec edent point pour l'extraction des lignes de cr^ ete (un point de cr^ ete est un passage par z ero de la d eriv ee de la courbure maximum le long de la direction maximum de courbure). Nous pr esentons des r esultats exp erimentaux sur l'extraction de lignes de cr^ etes sur des donn ees synth etiques et sur des donn ees m edicales r eelles.
منابع مشابه
A Multi-scale Approach for Crest Line Extraction in 3D Medical Images
Recently, we have shown that the differential properties of the surfaces represented by 3D volumic images can be recovered using their partial derivatives. For instance, the crest lines can be characterized by the first, second and third partial derivatives of the grey levrl function I(x,y,z) . This paper deals with the following points : the computation of the partial derivatives of an image c...
متن کاملExtraction of the zero-crossings of the curvature derivatives in volumic 3D medical images: a multi-scale approach
Volumic 3D images are now widely distributed in the medical field [HL79, TPRSS0, GHJ75, Aya93]. They are produced from various modalities such as Magnetic Resonance Imagery (MRI), Computed Tomography Imagery (CT), Nuclear Medicine Imagery (NMI) or Ultrasound Imagery (USI). data are represented by a discrete 3D grey level function I(i,j, k) where the high-contrast points (3D edge points) corresp...
متن کاملA 3D thin nets extraction method for medical imaging
This paper presents a powerful tool designed to extract characteristic lines, called 3D thin nets, from 3D volumetric images. 3D thin nets are the lines where the 3D grey level function is locally extremum in a given plane. Recently, we have shown that it is possible to characterize 2D thin nets as the crest lines of the image surface. This paper generalizes this approach to 3D data having thre...
متن کاملThe 3d Marching Lines Algorithm and Its Application to Crest Lines Extraction
This paper presents a powerful and general purpose tool designed to extract characteristic lines from 3D images . The algorithm, called Marching Lines, is inspired from the Marching Cubes algorithm, which is used to extract iso-value surfaces from 3D images. The Marching Lines algorithm extracts with sub-pixel accuracy the 3D lines corresponding to the intersection of two iso-surfaces from two ...
متن کاملExtraction and 3D Segmentation of Tumors-Based Unsupervised Clustering Techniques in Medical Images
Introduction The diagnosis and separation of cancerous tumors in medical images require accuracy, experience, and time, and it has always posed itself as a major challenge to the radiologists and physicians. Materials and Methods We Received 290 medical images composed of 120 mammographic images, LJPEG format, scanned in gray-scale with 50 microns size, 110 MRI images including of T1-Wighted, T...
متن کامل