Image segmentation method based on quaternion and fuzzy C-means clustering

The invention discloses an image segmentation method based on quaternion and fuzzy C-means clustering. The method includes the steps that an image to be segmented is converted into quaternion space, clustering amount is specified, a membership matrix U is initialized, the different degree between a current clustering center and pixels is measured through the defined quaternion distance, an objective function, a novel membership matrix and a clustering center are calculated, if the objective function is smaller than or equal to an iteration stopping threshold value, segmentation is completed, and a segmentation effect picture is output; if the objective function is larger than the iteration stopping threshold value, the novel membership matrix and the clustering center are used for calculating the quaternion distance and the objective function repeatedly until the objective function meets the iteration stopping condition, so that segmentation is completed. The image segmentation method based on quaternion spac