China 
Africa 
Explore chapters and articles related to this topic
Introduction
Published in Yu-Jin Zhang, A Selection of Image Analysis Techniques, 2023
Given an object in an image, the operation performed by the distance transform is to calculate the distance between each point in the object region and its closest point outside the region, and assign the distance value to the point. In other words, for a point in the object, the distance transform is defined as the closest distance between the point and the object boundary. More strictly, the distance transform can be defined as follows: Given a point set P, a subset B of P, and a distance function d(.,.) that satisfies the metric conditions (as in Sub-section 1.2.1), the distance transform of P is assigned a point p ∈ P the value: DT(p)=minq∈B{d(p,q)}
Introduction
Published in Vlad P. Shmerko, Svetlana N. Yanushkevich, Sergey Edward Lyshevski, Computer Arithmetics for Nanoelectronics, 2018
Vlad P. Shmerko, Svetlana N. Yanushkevich, Sergey Edward Lyshevski
The distance transform is used by many applications in order to map distances from feature points. The distance transform is also used in order to separate the adjoining features in biometric data and to identify primitives for the construction of new topological structures. The distance transform is normally applied to binary images and consists of calculating the distances between feature (colored) pixels and nonfeature (blank) pixels in the image; each pixel of the image is assigned a number that is the distance to the closest feature boundary.
Appropriate identification of age-related macular degeneration using OCT images
Published in Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization, 2021
Marwa Hani, Amine Ben Slama, Imen Zghal, Hedi Trabelsi
In our work, we have used region information about hyper-reflective zone detection in a simple way to refine watershed results. Then, a post-processing step is applied to obtain the optimal separating edges inside retinal region. To get initial detection of (AMD), we have applied the watershed algorithm by immersion (Tawse et al. 2014) on a gradient-weighted distance transform. The distance transform combines two image transformations: the geometric distance transform and the intensity gradients transform (Lin et al. 2003). It takes into consideration nuclear configurations and colour gradient information within nuclei. This distance can overcome the limitations of both geometrical and gradient transforms. On the one hand, the geometric distance is only interesting at dealing with regular shapes. On the other hand, the gradient transformation is sensitive to imaging noise, and usually results in over-segmentation (Li et al. 2011). The geometric distance D and the colour gradient transform ∆ are combined in a single representation given by the following expression (Grau et al. 2004):