l have an RGB image of dimension (224,224,3). l applied superpixel segmentation on it using SLIC algorithm.
我有一个维度(224,224,3)的RGB图像。 l使用SLIC算法对其进行超像素分割。
As follow :
如下 :
img= skimageIO.imread("first_image.jpeg")
print('img shape', img.shape) # (224,224,3)
segments_slic = slic(img, n_segments=1000, compactness=0.01, sigma=1) # Up to 1000 segments
segments_slic.shape
(224,224)
Number of returned segments are :
返回的段数是:
np.max(segments_slic)
Out[49]: 595
From 0 to 595. So, we have 596 superpixels (regions).
从0到595.所以,我们有596个超像素(区域)。
Let's take a look at segments_slic[0]
我们来看看segments_slic [0]
segments_slic[0]
Out[51]:
array([ 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,
1, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 3, 3, 3, 3, 3,
3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5, 5, 5,
5, 5, 5, 5, 6, 6, 6, 6, 6, 6, 6, 6, 6, 7, 7, 7, 7,
8, 8, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9, 9, 9, 9,
10, 10, 10, 10, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 12, 12,
12, 12, 12, 12, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 14, 14, 14,
14, 14, 14, 14, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 16, 16, 16,
16, 16, 16, 16, 17, 17, 17, 17, 17, 17, 17, 17, 17, 18, 18, 18, 18,
18, 18, 18, 18, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 19, 20, 20,
20, 20, 20, 20, 20, 20, 20, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21,
21, 21, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 23, 23, 23, 23, 23,
23, 23, 23, 23, 23, 23, 24, 24, 24, 24, 24, 24, 24, 25, 25, 25, 25,
25, 25, 25])
What l would like to get ?
我想得到什么?
for each superpixel region make two arrays as follow:
对于每个超像素区域,制作如下两个阵列:
1) Array : contain the indexes of the pixels belonging to the same superpixel.
1)数组:包含属于同一超像素的像素的索引。
For instance
superpixel_list[0] contains all the indexes of the pixels belonging to superpixel 0 .
superpixel_list [0]包含属于超像素0的像素的所有索引。
superpixel_list[400] contains all the indexes of the pixels belonging to superpixel 400
superpixel_list [400]包含属于超像素400的像素的所有索引
2)superpixel_pixel_values[0] : contains the pixel values (in RGB) of the pixels belonging to superpixel 0.
2)superpixel_pixel_values [0]:包含属于超像素0的像素的像素值(RGB)。
For instance, let's say that pixels 0, 24 , 29, 53 belongs to the superpixel 0. Then we get
例如,假设像素0,24,29,53属于超像素0.然后我们得到
superpixel[0]= [[223,118,33],[245,222,198],[98,17,255],[255,255,0]]# RGB values of pixels belonging to superpixel 0
What is the efficient/optimized way to do that ? (Because l have l dataset of images to loop over)
有效/优化的方法是什么? (因为我有l个图像数据集要循环)
EDIT-1
def sp_idx(s, index = True):
u = np.unique(s)
if index:
return [np.where(s == i) for i in u]
else:
return [s[s == i] for i in u]
#return [s[np.where(s == i)] for i in u] gives the same but is slower
superpixel_list = sp_idx(segments_slic)
superpixel = sp_idx(segments_slic, index = False)
In superpixel_list we are supposed to get a list containing the index of pixels belonging to the same superpixel. For instance superpixel_list[0] is supposed to get all the pixel indexes of the pixel affected to superpixel 0
在superpixel_list中,我们应该得到一个包含属于同一个超像素的像素索引的列表。例如,superpixel_list [0]应该获得受超像素0影响的像素的所有像素索引
however l get the following :
但我得到以下内容:
superpixel_list[0]
Out[73]:
(array([ 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2,
3, 3, 3, 3, 3, 3, 3, 4, 4, 4, 4, 4, 4, 4, 5, 5, 5,
5, 5, 5, 5, 5, 6, 6, 6, 6, 6, 6, 6, 6, 7, 7, 7, 7,
7, 7, 7, 8, 8, 8, 8, 8, 8, 8, 9, 9, 9, 9, 9, 9, 10,
10, 10, 10, 10, 11, 11, 11, 11, 12, 12, 12, 12, 13, 13, 13]),
array([0, 1, 2, 3, 4, 0, 1, 2, 3, 4, 5, 0, 1, 2, 3, 4, 5, 0, 1, 2, 3, 4, 5,
6, 0, 1, 2, 3, 4, 5, 6, 0, 1, 2, 3, 4, 5, 6, 7, 0, 1, 2, 3, 4, 5, 6,
7, 0, 1, 2, 3, 4, 5, 6, 0, 1, 2, 3, 4, 5, 6, 0, 1, 2, 3, 4, 5, 0, 1,
2, 3, 4, 0, 1, 2, 3, 0, 1, 2, 3, 0, 1, 2]))
Why two arrays ?
为什么两个阵列?
In superpixel[0] for instance we are supposed to get the RGB pixel values of each pixel affected to supepixel 0 as follow : for instance pixels 0, 24 , 29, 53 are affected to superpixel 0 then :
例如,在superpixel [0]中,我们应该得到受影响的每个像素的RGB像素值,如下所示:例如,像素0,24,29,53受到超像素0的影响,然后:
superpixel[0]= [[223,118,33],[245,222,198],[98,17,255],[255,255,0]]
However when l use your function l get the following :
但是当我使用你的函数时,我得到以下内容:
superpixel[0]
Out[79]:
array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0])
Thank you for your help
谢谢您的帮助
1 个解决方案
#1
1
Can be done using np.where and the resulting indices.
可以使用np.where和结果索引来完成。
def sp_idx(s, index = True):
u = np.unique(s)
return [np.where(s == i) for i in u]
superpixel_list = sp_idx(segments_slic)
superpixel = [img[idx] for idx in superpixel_list]
#1
1
Can be done using np.where and the resulting indices.
可以使用np.where和结果索引来完成。
def sp_idx(s, index = True):
u = np.unique(s)
return [np.where(s == i) for i in u]
superpixel_list = sp_idx(segments_slic)
superpixel = [img[idx] for idx in superpixel_list]