Connected Component Labeling Using Components Neighbors-Scan Labeling Approach
Akmal Rakhmadi, Nur Zuraifah Syazrah Othman, Abdullah Bade, Mohd Shafry Mohd Rahim and Ismail Mat Amin
DOI : 10.3844/jcssp.2010.1099.1107
Journal of Computer Science
Volume 6, Issue 10
Problem statement: Many approaches have been proposed in previous such as the classic sequential connected components labeling algorithm which is relies on two subsequent raster-scans of a binary image. This method produced good performance in terms of accuracy, but because of the implementation of the image processing systems now requires faster process of the computer, the speed of this technique’s process has become an important issue. Approach: A computational approach, called components neighbors-scan labeling algorithm for connected component labeling was presented in this study. This algorithm required scanning through an image only once to label connected components. The algorithm started by scanning from the head of the component’s group, before tracing all the components neighbors by using the main component’s information. This algorithm had desirable characteristics, it is simple while promoted accuracy and low time consuming. By using a table of components, this approach also gave other advantages as the information for the next higher process. Results: The approach had been tested with a collection of binary images. In practically all cases, the technique had successfully given the desired result. Averagely, from the results the algorithm increased the speed around 67.4% from the two times scanning method. Conclusion: Conclusion from the comparison with the previous method, the approach of components neighbors-scan for connected component labeling promoted speed, accuracy and simplicity. The results showed that the approach has a good performance in terms of accuracy, the time consumed and the simplicity of the algorithm.
© 2010 Akmal Rakhmadi, Nur Zuraifah Syazrah Othman, Abdullah Bade, Mohd Shafry Mohd Rahim and Ismail Mat Amin. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.