Colour saliency-based parameter optimisation for adaptive colour segmentation

Abstract

In this paper we present a parameter optimisation procedure that is designed to automatically initialise the number of clusters and the initial colour prototypes required by data space partitioning techniques. The proposed optimisation approach involves a colour saliency measure used in conjunction with a SOM classification procedure. For evaluation purposes, we have integrated the proposed initialisation technique in an unsupervised colour segmentation scheme based on K-Means clustering and the evaluation has been carried out in the context of the unsupervised segmentation of natural images.

Image segmentation based on the integration of colour–texture descriptors—A review

“…(3) Approaches that extract the colour and texture features on separate channels and then combine them in the segmentation process. These approaches can be further sub-categorised with respect to the strategy employed in the feature integration step as follows: (3.1) Region-based approaches that include: (a) split and merge [55][56][57][58] (3.2) Feature-based approaches that include statistical [78][79][80][81][82][83][84][85][86][88][89][90]92,95,97] and probabilistic segmentation schemes [98][99][100][101][102][105][106][107][108][109][110][111][112][113].…”

“…As a result, substantial research efforts have been devoted to develop robust initialisation schemes and to evaluate diverse algorithmic solutions to identify the optimal number of clusters in the input image. These issues have been specifically addressed in a recent colour-texture segmentation framework (referred to as CTex) [80,81], where colour and texture are investigated on separate channels. In this approach, the colour segmentation is the first major component of the proposed framework and involves the statistical analysis of data using multi-space colour representations.…”

Review of Graph, Medical and Color Image base Segmentation Techniques

“…The clustering methods [61] have been widely applied in the development of colour image segmentation algorithms due to their simplicity and low computational cost. Colour-texture segmentation framework (referred to as CTex) [62,63], where colour and texture are investigated on separate channels. The colour image segmentation involves filtering the input data using a Gradient-Boosted Forward and Backward (GB-FAB) anisotropic diffusion algorithm [64] that is applied to eliminate the influence of the image noise and improve the local colour coherence.…”