.jpg){kind=link}
Clustering of image by k-means algorithm
The source image is selected when Main.kt is starting
To display the image is used JPanel and BufferedImage in Image.kt
Clustered image is saved in root of project
IClusterizer.kt allows you to add others implementations of clustering
Cluster.kt and Pixel.kt just (nearly) data classes for clustering
KMeansClusterizer.kt - my implementation of k-means. Details of implementation are described in comments in class file.
Images layout in README by @alexander-lee description
| Source Image | 5-means Clustering (with 20 iterations) |
|---|---|
 |
 |
| 10-means Clustering (with 20 iterations) | 20-means Clustering (with 20 iterations) |
 |
 |