Efficient clustering approach for adaptive unsupervised colour image segmentation. Issue 10 (16th July 2019)

• Record Type: Journal Article
• Authors: Khan, Zubair; Yang, Jie; Zheng, Yuanjie
• Is Part Of: IET image processing. Volume 13:Issue 10(2019)
• Page Start: 1763
• Page End: 1772
• Publication Date: 2019-07-16
• Subjects: pattern clustering -- image colour analysis -- image segmentation -- computer vision -- image resolution; adaptive unsupervised colour image segmentation -- clustering‐based colour image segmentation approach -- colour image segmentation transforms image pixels -- image analysis -- computer vision applications -- image understanding -- pattern recognition -- adaptive unsupervised approach -- red–green–blue colour histogram search approach -- RGB histogram -- RGB triplet -- final segmented image -- unsupervised image segmentation algorithms -- image segmentation evaluation benchmarks -- efficient clustering approach; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2018.5976
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : This study proposes a clustering‐based colour image segmentation approach consisting of a novel initialisation technique. Colour image segmentation transforms image pixels into regions and a prerequisite for image analysis and computer vision applications. Therefore, colour image segmentation is considered one of the most important processes in image understanding and pattern recognition. This study presents an efficient and adaptive unsupervised approach based on bottom‐up red–green–blue (RGB) colour histogram search approach to achieve colour image segmentation. Firstly, the RGB histogram is processed through a double‐scan procedure to determine significant modes in each histogram. In the next step, each mode is processed through a bottom‐up histogram search approach, completing RGB triplet. The RGB triplets are utilised as the cluster centroids, clustering the pixels into regions and producing the final segmented image. The authors proposed method was compared with several other unsupervised image segmentation algorithms with an extensive experiment performed on various image segmentation evaluation benchmarks. Experimental results show that the proposed algorithm outperforms state‐of‐the‐art algorithms both in terms of features integrity and execution speed.