pyosv.ai.clustering 

 1from sklearn.cluster import KMeans
 2import numpy as np
 3
 4def pixel_clustering(img : np.ndarray, n_clusters : int = 3) -> np.ndarray:
 5    '''
 6        Image pixel clustering using Support Vector machine
 7        
 8        Parameters:
 9        -----------
10            - img : np.ndarray  
11                a WxHxB image, with width W, height H and B bands (channel last)
12            - n_clusters : int  
13                number of cluster to be identified
14        
15        Returns:
16        --------
17            - clusters : np.ndarray  
18                clustered pixels
19
20        Usage:
21        ------
22
23        ```python
24        import numpy as np  
25
26        img         = np.array(  
27            [[
28                [0.1, 0.2, 0.3],  
29                [0.4, 0.5, 0.6],  
30                [0.7, 0.8, 0.9]
31             ],
32             [
33                [1.1, 1.2, 1.3],  
34                [1.4, 1.5, 1.6],  
35                [1.7, 1.8, 1.9]
36             ]
37            ]  
38        ) 
39
40        # Making channels last
41        img = np.moveaxis(img, 0, -1)
42
43        clusters = pixel_clustering(img, n_clusters = 2)
44        ```
45        Output:
46        ------
47        ```
48        array([[0, 0, 0],
49               [0, 0, 1],
50               [1, 1, 1]], dtype=int32)
51       ```
52
53    '''
54
55    if len(img.shape) != 3:
56        raise Exception('Error: lenght of image shape must be 3 - (space, space, channels)')
57    
58    image_2D = img.reshape(img.shape[0]*img.shape[1], img.shape[2])
59    kmeans = KMeans(n_clusters = n_clusters, random_state = 0).fit(image_2D)
60    clusters = np.array(kmeans.labels_).reshape(img.shape[0], img.shape[1])
61    
62    return clusters
63
64def conv_clustering(): pass
def pixel_clustering(img: numpy.ndarray, n_clusters: int = 3) -> numpy.ndarray: 
 5def pixel_clustering(img : np.ndarray, n_clusters : int = 3) -> np.ndarray:
 6    '''
 7        Image pixel clustering using Support Vector machine
 8        
 9        Parameters:
10        -----------
11            - img : np.ndarray  
12                a WxHxB image, with width W, height H and B bands (channel last)
13            - n_clusters : int  
14                number of cluster to be identified
15        
16        Returns:
17        --------
18            - clusters : np.ndarray  
19                clustered pixels
20
21        Usage:
22        ------
23
24        ```python
25        import numpy as np  
26
27        img         = np.array(  
28            [[
29                [0.1, 0.2, 0.3],  
30                [0.4, 0.5, 0.6],  
31                [0.7, 0.8, 0.9]
32             ],
33             [
34                [1.1, 1.2, 1.3],  
35                [1.4, 1.5, 1.6],  
36                [1.7, 1.8, 1.9]
37             ]
38            ]  
39        ) 
40
41        # Making channels last
42        img = np.moveaxis(img, 0, -1)
43
44        clusters = pixel_clustering(img, n_clusters = 2)
45        ```
46        Output:
47        ------
48        ```
49        array([[0, 0, 0],
50               [0, 0, 1],
51               [1, 1, 1]], dtype=int32)
52       ```
53
54    '''
55
56    if len(img.shape) != 3:
57        raise Exception('Error: lenght of image shape must be 3 - (space, space, channels)')
58    
59    image_2D = img.reshape(img.shape[0]*img.shape[1], img.shape[2])
60    kmeans = KMeans(n_clusters = n_clusters, random_state = 0).fit(image_2D)
61    clusters = np.array(kmeans.labels_).reshape(img.shape[0], img.shape[1])
62    
63    return clusters

Image pixel clustering using Support Vector machine

Parameters:

 - img : np.ndarray  
     a WxHxB image, with width W, height H and B bands (channel last)
 - n_clusters : int  
     number of cluster to be identified

Returns:

 - clusters : np.ndarray  
     clustered pixels

Usage:

import numpy as np 

 
 img         = np.array( 

     [[
         [0.1, 0.2, 0.3], 

         [0.4, 0.5, 0.6], 

         [0.7, 0.8, 0.9]
      ],
      [
         [1.1, 1.2, 1.3], 

         [1.4, 1.5, 1.6], 

         [1.7, 1.8, 1.9]
      ]
     ] 

 ) 
 
 # Making channels last
 img = np.moveaxis(img, 0, -1)
 
 clusters = pixel_clustering(img, n_clusters = 2)

Output:

array([[0, 0, 0], [0, 0, 1], [1, 1, 1]], dtype=int32)

def conv_clustering(): 
65def conv_clustering(): pass