Class ComponentTreeGlobalImmersion<T>

java.lang.Object
imagingbook.common.mser.components.ComponentTree<T>
imagingbook.common.mser.components.ComponentTreeGlobalImmersion<T>
Type Parameters:
T - the type of properties to be attached to components
All Implemented Interfaces:
Iterable<Component<T>>

public class ComponentTreeGlobalImmersion<T> extends ComponentTree<T>

This class is a re-implementation of the "quasi-linear-time" component tree algorithm which is based on efficient, tree-based union finding as described in [1]. See Section 26.2.2 of [2] for a detailed description (Algs. 26.1 - 26.2). This algorithm is also used in the original MSER paper by Matas et al. [3].

[1] L. Vincent and P. Soille, "Watersheds in digital spaces: An efficient algorithm based on immersion simulations", IEEE Transactions on Pattern Analysis and Machine Intelligence 13(6), 583–598 (1991).
[2] W. Burger, M.J. Burge, Digital Image Processing – An Algorithmic Introduction, 3rd ed, Springer (2022).
[3] J. Matas, O. Chum, M. Urban, and T. Pajdla. Robust widebaseline stereo from maximally stable extremal regions. Image and Vision Computing 22(10), 761–767 (2004).

Version:
2022/11/19