| abstract |
The invention discloses an image segmentation method based on improved whale optimized fuzzy clustering, which mainly solves the problems of serious loss of image information after segmentation and long segmentation time in the prior art. The implementation steps are: 1. Input the image and obtain the gray levels of all pixels; 2. Select c cluster centers to divide the image into c categories; 3. Generate n whales, each whale has a c-dimensional vector, representing A possible solution of a group of cluster centers; 4. Use the reciprocal of the fast fuzzy C-means clustering objective function as the fitness value to search for the optimal cluster center; 5. A group of clusters corresponding to the searched maximum fitness value The center implements image segmentation, classifies the pixels whose gray levels are in the same membership interval into one category, and outputs the segmented image. The invention improves the image segmentation effect by combining the optimization result with the fuzzy clustering image segmentation, and can be used for target detection, video monitoring and medical imaging. |