Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Hybrid object labelling in digital images
Martín-Herrero J. Machine Vision and Applications18 (1):1-15,2007.Type:Article
Date Reviewed: Dec 21 2007

Labelling is a computer-vision task that identifies all the connected pixels of an object with the same label. It is often followed by the feature extraction, characterization, for the further classification of objects.

Labelling algorithms basically differ in either a recursive or iterative nature. The classical recursive algorithm fits well for time constrained applications because it doesn’t only allow real-time labelling, but characterization at the same time. Both recursive and iterative algorithms use the stack--either the system stack for recursive calls, or the custom stack to allocate data structures. The paper focuses on the risk of stack overflow in labelling algorithms--a fact that can have drastic consequences in real-time applications. The author studies the performance of a very simple labelling algorithm that is a denoted hybrid algorithm, because it combines recursive and iterative approaches, and with the aim of reducing recursivity while having its advantages at the same time.

The paper is easy to read. It exhaustively provides the tests results, which include comparisons of stack use, and time performance versus classical recursive and iterative solutions. Tests on synthetic images of different sizes, both purely random and block structured with several densities, allow probing the robustness of the algorithm. The author also shows the test results of real-time, two-dimensional (2D) and three-dimensional (3D), machine vision (MV) applications. Furthermore, the paper demonstrates how to apply the algorithm to detect spanning objects.

After reading the paper, the reader is persuaded to either choose or replace any labelling implementation with this labelling hybrid algorithm.

Reviewer:  Marma Abasolo Review #: CR135047
Bookmark and Share
 
Object Recognition (I.4.8 ... )
 
 
Feature Evaluation And Selection (I.5.2 ... )
 
 
Graph Labeling (G.2.2 ... )
 
 
Recursive Function Theory (F.4.1 ... )
 
 
Design Methodology (I.5.2 )
 
 
Graph Theory (G.2.2 )
 
  more  
Would you recommend this review?
yes
no
Other reviews under "Object Recognition": Date
On view likelihood and stability
Weinshall D., Werman M. IEEE Transactions on Pattern Analysis and Machine Intelligence 19(2): 97-108, 1997. Type: Article
Feb 1 1999
Junctions
Parida L., Geiger D., Hummel R. IEEE Transactions on Pattern Analysis and Machine Intelligence 20(7): 687-698, 1998. Type: Article
Apr 1 1999
Topological recognition of polyhedral objects from multiple views
Laurentini A. Artificial Intelligence 127(1): 31-55, 2001. Type: Article
Apr 17 2002
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy