Image Segmentation Using Superpixel Based Split and Merge Method

Authors

  • Loay . K.Abood Department of Computer Science College of Science, University of Baghdad, Baghdad, Iraq
  • Raad. A.Mohammed Department of Computer Science College of Science, University of Baghdad, Baghdad, Iraq

Keywords:

super pixels, image Segmentation, graph cut

Abstract

A super pixel can be defined as a group of pixels, which have similar characteristics, which can be very helpful for image segmentation. It is generally color based segmentation as well as other features like texture, statistics…etc .There are many algorithms available to segment super pixels like Simple Linear Iterative Clustering (SLIC) super pixels and Density-Based Spatial Clustering of Application with Noise (DBSCAN). SLIC algorithm essentially relay on choosing N random or regular seeds points covering the used image for segmentation. In this paper Split and Merge algorithm was used instead to overcome determination the seed point's location and numbers as well as other used parameters. The overall results were better from the SLIC method depending on single threshold, which control the segments number needed (like 0.2) to accomplish the task.

Downloads

Download data is not yet available.

Downloads

Published

2023-07-02

Issue

Section

Computer Science

How to Cite

Image Segmentation Using Superpixel Based Split and Merge Method. (2023). Iraqi Journal of Science, 56(1A), 233-237. https://ijs.uobaghdad.edu.iq/index.php/eijs/article/view/10718

Similar Articles

81-90 of 656

You may also start an advanced similarity search for this article.