Hambali, Hamirul ’Aini and Syed Abdullah, Sharifah Lailee and Jamil, Nursuriati and Harun, Hazaruddin (2016) Intelligent segmentation of fruit images using an integrated thresholding and adaptive K-means method (TSNKM). Jurnal Teknologi, 78 (6-5). pp. 13-20. ISSN 0127-9696

Preview

PDF Download (445kB) | Preview

Abstract

Recent years, vision-based fruit grading system is gaining importance in fruit classification process.In developing the fruit grading system, image segmentation is required for analyzing the fruit objects automatically.Image segmentation is a process that divides a digital image into separate regions with the aim to obtain only the interest objects and remove the background. Currently, there are several segmentation techniques which have been used in object identification such as thresholding and clustering techniques.However, the conventional techniques have difficulties in segmenting fruit images which captured under natural illumination due to the existence of non-uniform illumination on the object surface.The presence of different illuminations influences the appearance of the interest objects and thus misleads the object analysis.Therefore, this research has produced an innovative segmentation algorithm for fruit images which is able to increase the segmentation accuracy.The developed algorithm is an integration of modified thresholding and adaptive K-means method.The integration of both methods is required to increase the segmentation accuracy for fruits images with different surface colour.The results showed that the innovative method is able to segment the fruits images with high accuracy value.

Item Type:	Article
Uncontrolled Keywords:	Segmentation, thresholding, K-means, Fuzzy C-means, active contour, natural illumination
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:	School of Computing
Depositing User:	Mdm. Hamirul `Aini Hambali
Date Deposited:	18 Jan 2017 03:06
Last Modified:	18 Jan 2017 03:06
URI:	https://repo.uum.edu.my/id/eprint/20635

Actions (login required)

View Item