Improved genetic FCM algorithm for color image segmentation

An improved genetic fuzzy c-means clustering(FCM) algorithm is proposed for color image segmentation in the paper. The first component of color feature set discovered by Ohta is chosen as the one-dimensional eigenvector and the mapping from pixel space to eigenvector space is used here for modifying the object function in order to lower the computational complexity. Feature distance which is applied to any structure of eigenvector space is used here instead of Euclidian distance to reduce the influence caused by structure of eigenvector space. FCM optimization is introduced to genetic FCM algorithm to accelerate the searching speed. Experiments show that the algorithm has better effect on color image segmentation and low computational complexity.