-
[show abstract]
[hide abstract]
ABSTRACT: Thin nets are the lines where the grey level function is locally extremum in a given direction. Recently, we have shown that it is possible to characterize the thin nets using differential properties of the image surface. However, the method failed when these structures present different widths. In this paper we show that, the extraction process of the thin nets, having different width, requires a multi-scale analysis of the image. To design the fusion process of the multi-scale information, we will study the behavior of the differential properties of the image surface, in particular the curvatures, in scale space. We illustrate the efficiency of the proposed multi-scale approach by extracting roads of different widths in satellite images.
04/2006: pages 361-364;
-
[show abstract]
[hide abstract]
ABSTRACT: Thin nets are the lines where the grey level function is locally extremum in a given direction. Recently, we have shown that it is possible to characterize the thin nets using differential properties of the image surface. However, the method failed when these structures present different widths. In this paper we show that the extraction process of the thin nets, having different width, requires a multi-scale analysis of the image. To design the fusion process of the multi-scale information, we will study the behavior of the differential properties of the image surface, in particular the curvatures, in scale space. We illustrate the efficiency of the proposed multi-scale approach by extracting roads and blood vessels of different widths in satellite and medical images.
Computer Vision and Image Understanding. 01/1999;
-
[show abstract]
[hide abstract]
ABSTRACT: In this paper, we describe a new approach for extracting thin nets in 2D grey level images. The key point is to model thin nets as the crest lines of the image surface. Crest lines are the lines where one of the two principal curvatures is locally extremal. We define these lines using first, second and third derivatives of the image. We compute the image derivatives using recursive filters approximating the Gaussian filter and its derivatives. This paper presents an algorithm to extract thin nets from 2D images and we apply this method to the extraction of roads in satellite images and blood vessels in medical images. 1 Introduction In many images, Thin Nets (TN) correspond to important features [6] [7]. For instance, in aerial and medical images, TN are attached respectively to roads and blood vessels. TN are formed by the points where the grey-level is locally extremum in a given direction. This direction is the normal to the curve traced by the TN at this point. Classic edge detect...
04/1998;
-
[show abstract]
[hide abstract]
ABSTRACT: We describe a new approach for extracting thin nets in grey-level images. The key point of our approach is to model thin nets as the crest lines of the image surface. These lines are defined using first, second and third derivatives of the image. We apply this approach to the extraction of blood vessels in medical images. 1 Introduction In many images, Thin Nets (TN) correspond to important features [3]. For instance, in medical images, TN are attached to blood vessels. TN are formed by the points where the grey level is locally extremum in a given direction. This direction is the normal to the curve traced by the TN at this point. Here, we propose a new method to detect TN using differential geometry. An important point of our approach is its ability to identify TN by crest lines on the surface defined by the image [5][4]. We have tested this method on medical images in which the goal is to extract blood vessels. The quality of the results obtained clearly shows the flexibility and t...
04/1998;
-
[show abstract]
[hide abstract]
ABSTRACT: 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 focally
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 three principal
curvatures of the hypersurface defined by the 3D volumetric image. Using
differential properties of image hypersurfaces, we explain that 3D thin
nets can be extracted by intersecting of two surfaces, one corresponding
to the maximization of maximum curvature in its associated direction,
and the other one to the maximization of medium curvature in its
associated direction. We apply this approach to the extraction of blood
vessels in 3D medical images
Pattern Recognition, 1996., Proceedings of the 13th International Conference on; 09/1996
-
[show abstract]
[hide abstract]
ABSTRACT: We describe a new approach for extracting crest lines and thin nets. The key point of our approach is to model thin nets as the crest lines of the image surface. Crest lines are the lines where one of the two principal curvatures is locally extremal. We define these lines using first, second and third derivatives of the image. We compute the image derivatives using recursive filters approximating the Gaussian filter and its derivatives. Using an adapted scale factor, we apply this approach to the extraction of roads in satellite data and blood vessels in medical images. We also apply this method to the extraction of the crest lines in depth maps of human faces
Image Processing, 1995. Proceedings., International Conference on; 11/1995
