Image Segmentation Based on Fast High Dimensional Characteristic Clustering Using Fusion of Classifiers

Abstract

Image segmentation can be viewed as a classification problem of high dimensional characteristic data including grayscale (color), spatial constraint, etc. In this paper, a new image segmentation algorithm based on fast high-dimensional characteristic clustering using fusion of classifiers is proposed. In the algorithm, classification of high-dimensional characteristic data is divided into multiple low-dimensional characteristic data classifications such as optimal fuzzy classification based on grayscale (color) characteristic and statistical classification based on spatial constraint. The classification results of different classifiers are integrated to obtain the final image segmentation results using fusion of classifiers. The experimental results show that, compared with other image classification algorithms, the proposed algorithm in this paper has better segmentation performance and greatly improves the calculation speed, ensuring the calculation simplicity and effectiveness of segmentation algorithm to the largest extent.