<abstract xml:lang="en" xmlns="http://www.w3.org/1999/xhtml"><p>Image segmentation is first and very important step in image analysis. The main idea of image segmentation is to simplify and change image into easier and meaningful form to analyze. Image segmentation is process, which locate objects in image. Many segmentation algorithms have been created for different applications. The algorithms are used in traffic applications, army applications, web applications, medical applications, studying and many others. In present time, do not exist restful objective methods to evaluate segmentation algorithms. This paper presents evaluation criterion based on measurement of precision of boundary segmentation. Moreover, the automatic segmentation algorithms in comparison with human segmentation results were tested. Four most used image segmentation algorithms, namely, Efficient graph based, K-means, Mean shift and Belief propagation are compared by designed criterion. The criterion computes three evaluation parameters like precision, recall and <italic>F</italic><sub>1</sub> and the results are presented in the tables and graphs at the end of the paper.</p></abstract>

Image segmentation is first and very important step in image analysis. The main idea of image segmentation is to simplify and change image into easier and meaningful form to analyze. Image segmentation is process, which locate objects in image. Many segmentation algorithms have been created for different applications. The algorithms are used in traffic applications, army applications, web applications, medical applications, studying and many others. In present time, do not exist restful objective methods to evaluate segmentation algorithms. This paper presents evaluation criterion based on measurement of precision of boundary segmentation. Moreover, the automatic segmentation algorithms in comparison with human segmentation results were tested. Four most used image segmentation algorithms, namely, Efficient graph based, K-means, Mean shift and Belief propagation are compared by designed criterion. The criterion computes three evaluation parameters like precision, recall and F1 and the results are presented in the tables and graphs at the end of the paper.

The Evaluation Criterion for Color Image Segmentation Algorithms

Department of Telecommunications and Multimedia, University of Žilina, Univerzitná, 010 26 Žilina, Slovakia

Journal of Electrical Engineering

{"article-title":"The Evaluation Criterion for Color Image Segmentation Algorithms"}

Image segmentation is first and very important step in image analysis. The main idea of image segmentation is to simplify and change image into easier and...

The Evaluation Criterion for Color Image Segmentation Algorithms

Published Online: Feb 22, 2012

Page range: 13 - 20

DOI: https://doi.org/10.2478/v10187-012-0002-1

Keywords
image, segmentation, algorithm, evaluation criterion

This content is open access.

The Evaluation Criterion for Color Image Segmentation Algorithms

Published Online: Feb 22, 2012

Page range: 13 - 20

DOI: https://doi.org/10.2478/v10187-012-0002-1

Keywordsimage, segmentation, algorithm, evaluation criterion

This content is open access.

Keywords
image, segmentation, algorithm, evaluation criterion