Automatic segmentation of fruits in CIELuv color space image using hill climbing optimization and fuzzy C-Means clustering

Abstract

In this paper, a novel method for the segmentation and extraction of natural fruits using Hill climbing (HC) optimization and Modified Fuzzy C-Means (MFCM) clustering algorithm is proposed. The intensity and color information is highly correlated in RGB color images. The segmentation in RGB color space does not produce the meaningful outcome for the segmentation and information retrieval. Many authors have proposed different color space for the segmentation and retrieval of information. In this color based segmentation technique, RGB color images had transformed into perceptually uniform, device independent CIELuv color space for the efficient segmentation. Then for the CIELuv image, the color histogram had generated and computed. This color histogram acts as a search space and has a number of bins. In this work, Hill climbing (HC) optimization had applied for the identification of best image pixels (peaks) which correspond to the initial number of seeds or clusters for the segmentation process. These initial seeds had applied to MFCM for the segmentation of fruits in CIELuv color images. The experimental result had compared with the segmentation process in RGB color space to demonstrate the efficiency of the proposed approach.