Distance transformations in digital images

By Gunilla Borgefors. 1986

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Shape-aware Instance Segmentation (2016, methodology): “To this end, instead of directly inferring a binary mask, which would only represent the visible portion of the object, we propose to construct a pixel-wise, multi-valued map encoding the shape of the complete object by relying on the concept of distance transform [2].”
A Linear Time Algorithm for Computing Exact Euclidean Distance Transforms of Binary Images in Arbitrary Dimensions (2003, unknown): “Algorithms that require OðNÞ time for computing the DT of a k-D image for the L1, L1, and chamfer (a weighted L1 metric that provides an approximation to theL2 metric) metrics with a two-pass sequential raster scan are well-known [4], [5], [25], [26], [34], [35], [48].”
ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data (2019, background): “The third is the distance transform2 [7] of the segmentation mask.”
The watershed concept and its use in segmentation : a brief history (2012, unknown): “Borgefors introduced chamfer distances [10].”
Enhanced Local Texture Feature Sets for Face Recognition Under Difficult Lighting Conditions (2007, methodology): “12 (top) shows the extent to which nearest neighbour based LBP face recognition can be improved by combining three of the enhancements proposed here: using preprocessing (PP); replacing LBP with LTP; and replacing local his-togramming and the χ 2 histogram distance with the Distance Transform based ”
2D Euclidean distance transform algorithms: A comparative survey (2008, methodology): “—Shape matching [Borgefors 1986; Paglieroni 1992a; Liu and Srinath 1990; You et al. 1995].”
Spectral–Spatial Classification of Hyperspectral Imagery Based on Partitional Clustering Techniques (2009, unknown): “3 (that are 8- and 16-neighborhoods of a pixel, called Chamfer neighborhoods [62]).”
Segmentation and Tracking of Multiple Humans in Crowded Environments (2008, methodology): “First, the Canny edge detector is applied to the foreground region oftheinputimage.Adistancetransformation[ 1 ]iscomputed on the edge map.”
Learning OpenCV 3: Computer Vision in C++ with the OpenCV Library (2016, methodology): “This is the “original” method developed by Borgefors (1986) [Borgefors86].”
Bayesian human segmentation in crowded situations (2003, methodology): “A distance transformation [1] is then computed on the edge map.”
Efficient matching of pictorial structures (2000, methodology): “The resulting approximation has a fixed percentage error (see [3]).”
Segmentation of confocal microscope images of cell nuclei in thick tissue sections (1999, methodology): “5) was determined using a modified vector distance transform (Mullikin, 1992) to avoid both the expensive computational load of the Euclidean distance transform (Castleman, 1979b) and the errors associated with the use of the cityblock distance transform (Borgefors, 1986).”