Gradual Color Clustering Elimination as a Novel and Efficient method for Outdoor Image Segmentation

Abstract

One of the color reduction methods is color clustering, which has been applied for segmentation.
Nonetheless, it has not been an appropriate method due to the automatically images change by
luminance effects and color/texture variety. Hence, it can be done by improving the usual color
clustering methods called customizing segmentation methods. This study focuses on customizing the
color clustering methods for segmentation and object recognition in the outdoor images by utilizing a
multi-phase procedure through a multi-resolution platform, based on self-organizing neural network,
called gradual color Cluster Elimination (GCCE). The proposed method has been evaluated on
outdoor images dataset namely BSDS and the results have been compared to PRI, NPR, and GCE
statistical metrics of the latest segmentation methods which demonstrated that the proposed method
has a satisfactory performance for the segmentation of the outdoor scenes.