A New Minimum Barrier Distance for Multivariate Images with Applications to Salient Object Detection, Shortest Path Finding, and Segmentation
From LRDE
- Authors
- Minh Ôn Vũ Ngoc, Nicolas Boutry, Jonathan Fabrizio, Thierry Géraud
- Journal
- Computer Vision and Image Understanding
- Type
- article
- Projects
- Olena
- Keywords
- Image
- Date
- 2020-06-02
Abstract
Distance transforms and the saliency maps they induce are widely used in image processing, computer vision, and pattern recognition. One of the most commonly used distance transform is the geodesic one. Unfortunately, this distance does not always achieve satisfying results on noisy or blurred images. Recently, a new (pseudo-)distance, called the minimum barrier distance (MBD), more robust to pixel variations, has been introduced. Some years after, Géraud et al. have proposed a good and fast-to compute approximation of this distance: the Dahu pseudo-distance. Since this distance was initially developped for grayscale images, we propose here an extension of this transform to multivariate images; we call it vectorial Dahu pseudo-distance. An efficient way to compute it is provided in this paper. Besides, we provide benchmarks demonstrating how much the vectorial Dahu pseudo-distance is more robust and competitive compared to other MB-based distances, which shows how much this distance is promising for salient object detection, shortest path finding, and object segmentation.
Documents
Bibtex (lrde.bib)
@Article{ movn.20.cviu, author = {Minh {\^On V\~{u} Ng\d{o}c} and Nicolas Boutry and Jonathan Fabrizio and Thierry G\'eraud}, title = {A New Minimum Barrier Distance for Multivariate Images with Applications to Salient Object Detection, Shortest Path Finding, and Segmentation}, journal = {Computer Vision and Image Understanding}, year = {2020}, month = aug, volume = {197--198}, doi = {10.1016/j.cviu.2020.102993}, abstract = {Distance transforms and the saliency maps they induce are widely used in image processing, computer vision, and pattern recognition. One of the most commonly used distance transform is the geodesic one. Unfortunately, this distance does not always achieve satisfying results on noisy or blurred images. Recently, a new (pseudo-)distance, called the minimum barrier distance (MBD), more robust to pixel variations, has been introduced. Some years after, G\'eraud et al. have proposed a good and fast-to compute approximation of this distance: the Dahu pseudo-distance. Since this distance was initially developped for grayscale images, we propose here an extension of this transform to multivariate images; we call it vectorial Dahu pseudo-distance. An efficient way to compute it is provided in this paper. Besides, we provide benchmarks demonstrating how much the vectorial Dahu pseudo-distance is more robust and competitive compared to other MB-based distances, which shows how much this distance is promising for salient object detection, shortest path finding, and object segmentation.} }