A Region-Based Image Segmentation by Watershed Partition and DCT Energy Compaction

An image segmentation approach by improved watershed partition and DCT energy compaction has been proposed in this paper. The proposed energy compaction, which expresses the local texture of an image area, is derived by exploiting the discrete cosine transform. The algorithm is a hybrid segmentation...

Full description

Saved in:
Bibliographic Details
Published in2011 Eighth International Conference Computer Graphics, Imaging and Visualization pp. 131 - 135
Main Authors Chi-Man Pun, Ning-Yu An, Miao Cheng
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.08.2011
Subjects
Algorithm design and analysis
Compaction
cosine transform
Discrete cosine transforms
energy compaction
Image segmentation
Merging
watershed
Online AccessGet full text

Cover

More Information
Summary:An image segmentation approach by improved watershed partition and DCT energy compaction has been proposed in this paper. The proposed energy compaction, which expresses the local texture of an image area, is derived by exploiting the discrete cosine transform. The algorithm is a hybrid segmentation technique which is composed of three stages. First, the watershed transform is utilized by preprocessing techniques: edge detection and marker in order to partition the image into several small disjoint patches, while the three features: region size, mean and variance are used to calculate region energy for combination. Then in the second merging stage, the DCT transform is used for energy compaction which is a criterion for texture comparison and region merging. Finally the image can be segmented into several partitions. The obtained results show good segmentation robustness and efficiency, when compared to other state of the art image segmentation algorithms.
DOI:10.1109/CGIV.2011.27