Paper Details: Downloads: 1260
Serial Number: P1151553461
Title: Efficient Color Image Segmentation relying on Binary Image Interest Points
Authors: E aryby Nada and Aitfares Wassima and Regragui Fakhita
Abstract: The present work proposes a new method for color image segmentation. This approach is based on the calculation of the interest points from the binary image to extract the binary mask used to separate the object from its background. The binarization of the image is done by using the local Otsu threshold method to reduce the effect of heterogeneity thus improving the object segmentation. To evaluate the effectiveness of the proposed method, we compared the obtained results of the object segmentation based on our approach with the results of the grab cut method. Experimental results show the robustness of our approach for detecting objects in real images.
Keywords: Color image segmentation, Lab color space, heterogeneity, local Otsu thresholding, interest points, morphological processing
Journal/Conference: International Journal of Graphics, Vision and Image Processing
Volume: 16
Issue: 2
Submission Date: 12/28/2015 12:00:00 AM
Review Date: 3/28/2016 12:00:00 AM
Publishing Date: 5/2/2016 12:00:00 AM
Article Downloads: 1260
Download:

Facebook