12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455 |
- #!/usr/bin/env python
- '''
- K-means clusterization sample.
- Usage:
- kmeans.py
- Keyboard shortcuts:
- ESC - exit
- space - generate new distribution
- '''
- # Python 2/3 compatibility
- from __future__ import print_function
- import numpy as np
- import cv2 as cv
- from gaussian_mix import make_gaussians
- def main():
- cluster_n = 5
- img_size = 512
- # generating bright palette
- colors = np.zeros((1, cluster_n, 3), np.uint8)
- colors[0,:] = 255
- colors[0,:,0] = np.arange(0, 180, 180.0/cluster_n)
- colors = cv.cvtColor(colors, cv.COLOR_HSV2BGR)[0]
- while True:
- print('sampling distributions...')
- points, _ = make_gaussians(cluster_n, img_size)
- term_crit = (cv.TERM_CRITERIA_EPS, 30, 0.1)
- _ret, labels, _centers = cv.kmeans(points, cluster_n, None, term_crit, 10, 0)
- img = np.zeros((img_size, img_size, 3), np.uint8)
- for (x, y), label in zip(np.int32(points), labels.ravel()):
- c = list(map(int, colors[label]))
- cv.circle(img, (x, y), 1, c, -1)
- cv.imshow('kmeans', img)
- ch = cv.waitKey(0)
- if ch == 27:
- break
- print('Done')
- if __name__ == '__main__':
- print(__doc__)
- main()
- cv.destroyAllWindows()
|