Re: GLCM calculation using scikit-learn. Error when using greycomatrix

[ioannisgkan259]

Thank you very much for your quick reply, i appreciate it. It seems you 
always have a good answer to give me.

Ioannis 

On Sunday, 2 October 2016 03:41:10 UTC+1, Juan Nunez-Iglesias wrote:
>
> Hi Ioannis,
>
> So, the size of the GLCM for a 32-bit image that uses all possible 
> intensity levels is 2**64, way too big to fit in memory. The key is:
>
> - rescale the image to be in [0, max_value-1] (for some integer max_value 
> that you determine)
> - call greycomatrix on this rescaled image with the keyword argument 
> `levels=max_value`.
>
> That should work. However, note that the size of the matrix will be `4 * 
> max_value ** 2` or `8 * max_value ** 2`, and multiple temporary matrices of 
> the same size might be created for processing. So, you need to make sure 
> that your computer has enough memory for that. Typically, this will require 
> your image to have much fewer levels than even 16-bit.
>
> Additionally, I think this feature requires the development version of 
> scikit-image. You can install that with "pip install git+
> https://github.com/scikit-image/scikit-image.git;. (I think that's the 
> command anyway...)
>
> Juan.
>
> On Sun, Oct 2, 2016 at 6:33 AM,  
> wrote:
>
>> Hello Juan,
>>
>> I want to thank you for your quick reply to my question the last time and 
>> i do not want to bother you with my questions. I always try to solve my 
>> problems before posting my question here. Sometime, i just cannot figure 
>> out how to fix the error.
>>
>> What i want to do is to create a GLCM image using different  grey levels 
>> (8-bit, 16-bit and 32-bit) and compare the results. The type of the image i 
>> am using is a float32. I used the '' rescale intensity'' as you suggested, 
>> i converted the image to 8-bit and it worked. When i try to rescale the 
>> image in 16-bit and 32-bit for some reason it doesn't work.
>>
>> i get the error:
>>
>>
>> AssertionErrorTraceback (most recent call 
>> last) in () 22  23 for j 
>> in xrange(img.shape[1] ):---> 24 glcm = greycomatrix(rescale, [1], 
>> [0],  symmetric = True, normed = True ) 25  26 
>> testraster1[i,j] = greycoprops(glcm, 'contrast')
>> C:\Anaconda2\lib\site-packages\skimage\feature\texture.pyc in 
>> greycomatrix(image, distances, angles, levels, symmetric, normed)101 
>> image = np.ascontiguousarray(image)102 assert image.min() >= 0--> 
>> 103 assert image.max() < levels104 image = 
>> image.astype(np.uint8)105 distances = 
>> np.ascontiguousarray(distances, dtype=np.float64)
>> AssertionError: 
>>
>> It seems that i am allowed only to use an 8-bit image for the GLCM 
>> calculation.
>> So, does this code in scikit-image work only for 8-bit images? is there 
>> another way to solve my problem?
>> I will keep trying to find a solution but if you have an idea on how to 
>> solve this problem, please tell me
>>
>> Thank you in advance
>> Ioannis
>>
>>
>>
>> On Monday, 19 September 2016 15:41:07 UTC+1, ioannis...@gmail.com wrote:
>>>
>>> Thank you very much Juan for your quick reply.
>>> That was helpful :)
>>>
>>> Ioannis
>>>
>>> On Monday, 19 September 2016 01:03:45 UTC+1, Juan Nunez-Iglesias wrote:

 Hi Ioannis,

 Unfortunately the levels keyword is used as a hint to the function 
 about the number of levels when the image is uint16, because the possible 
 number of levels is huge. But if you want to convert the image to those 
 levels, you have to do it manually. I suggest you look at the 
 "rescale_intensity" function:


 http://scikit-image.org/docs/dev/api/skimage.exposure.html#skimage.exposure.rescale_intensity

 and process your image before passing it to the glcm function.

 I hope this helps! Keep pinging if you have more questions. =)

 Juan.

 On Sun, Sep 18, 2016 at 4:57 AM,  wrote:

> Hello everyone,
>
> I am using a SAR image (16-bit) and trying to implement GLCM algorithm 
> using sciki-learn. When trying to calculate the GLCM using greycomatrix i 
> get the following error: 
>
> assert image.max() < levels. It says that the maximum value of the image 
> intensity must be less than the number of grey levels.
> Because the SAR image is really big, i want to reduce the calculation 
> time by reducing the levels to 8. 
> Even if i remove the parameter 'level=8' when using greycomatrix, still 
> gives me the same error
>
> My code is the following:
>
> from skimage.feature import greycomatrix, greycoprops
> import numpy as np
> from skimage import data
> import rasterio
>
> path = 'C:\Users\GLCM_implementation\glasgow.tif'
>
> with rasterio.open(path, 'r') as src:
> import_file = src.read()
> img = import_file[0,:,:] #i need only the two dimentions (height, 
> width)
> print img.shape

Re: GLCM calculation using scikit-learn. Error when using greycomatrix

[Juan Nunez-Iglesias]

Hi Ioannis,

So, the size of the GLCM for a 32-bit image that uses all possible
intensity levels is 2**64, way too big to fit in memory. The key is:

- rescale the image to be in [0, max_value-1] (for some integer max_value
that you determine)
- call greycomatrix on this rescaled image with the keyword argument
`levels=max_value`.

That should work. However, note that the size of the matrix will be `4 *
max_value ** 2` or `8 * max_value ** 2`, and multiple temporary matrices of
the same size might be created for processing. So, you need to make sure
that your computer has enough memory for that. Typically, this will require
your image to have much fewer levels than even 16-bit.

Additionally, I think this feature requires the development version of
scikit-image. You can install that with "pip install git+
https://github.com/scikit-image/scikit-image.git;. (I think that's the
command anyway...)

Juan.

On Sun, Oct 2, 2016 at 6:33 AM,  wrote:

> Hello Juan,
>
> I want to thank you for your quick reply to my question the last time and
> i do not want to bother you with my questions. I always try to solve my
> problems before posting my question here. Sometime, i just cannot figure
> out how to fix the error.
>
> What i want to do is to create a GLCM image using different  grey levels
> (8-bit, 16-bit and 32-bit) and compare the results. The type of the image i
> am using is a float32. I used the '' rescale intensity'' as you suggested,
> i converted the image to 8-bit and it worked. When i try to rescale the
> image in 16-bit and 32-bit for some reason it doesn't work.
>
> i get the error:
>
>
> AssertionErrorTraceback (most recent call 
> last) in () 22  23 for j 
> in xrange(img.shape[1] ):---> 24 glcm = greycomatrix(rescale, [1], 
> [0],  symmetric = True, normed = True ) 25  26 
> testraster1[i,j] = greycoprops(glcm, 'contrast')
> C:\Anaconda2\lib\site-packages\skimage\feature\texture.pyc in 
> greycomatrix(image, distances, angles, levels, symmetric, normed)101 
> image = np.ascontiguousarray(image)102 assert image.min() >= 0--> 103 
> assert image.max() < levels104 image = image.astype(np.uint8)
> 105 distances = np.ascontiguousarray(distances, dtype=np.float64)
> AssertionError:
>
> It seems that i am allowed only to use an 8-bit image for the GLCM 
> calculation.
> So, does this code in scikit-image work only for 8-bit images? is there 
> another way to solve my problem?
> I will keep trying to find a solution but if you have an idea on how to solve 
> this problem, please tell me
>
> Thank you in advance
> Ioannis
>
>
>
> On Monday, 19 September 2016 15:41:07 UTC+1, ioannis...@gmail.com wrote:
>>
>> Thank you very much Juan for your quick reply.
>> That was helpful :)
>>
>> Ioannis
>>
>> On Monday, 19 September 2016 01:03:45 UTC+1, Juan Nunez-Iglesias wrote:
>>>
>>> Hi Ioannis,
>>>
>>> Unfortunately the levels keyword is used as a hint to the function about
>>> the number of levels when the image is uint16, because the possible number
>>> of levels is huge. But if you want to convert the image to those levels,
>>> you have to do it manually. I suggest you look at the "rescale_intensity"
>>> function:
>>>
>>> http://scikit-image.org/docs/dev/api/skimage.exposure.html#s
>>> kimage.exposure.rescale_intensity
>>>
>>> and process your image before passing it to the glcm function.
>>>
>>> I hope this helps! Keep pinging if you have more questions. =)
>>>
>>> Juan.
>>>
>>> On Sun, Sep 18, 2016 at 4:57 AM,  wrote:
>>>
 Hello everyone,

 I am using a SAR image (16-bit) and trying to implement GLCM algorithm
 using sciki-learn. When trying to calculate the GLCM using greycomatrix i
 get the following error:

 assert image.max() < levels. It says that the maximum value of the image 
 intensity must be less than the number of grey levels.
 Because the SAR image is really big, i want to reduce the calculation time 
 by reducing the levels to 8.
 Even if i remove the parameter 'level=8' when using greycomatrix, still 
 gives me the same error

 My code is the following:

 from skimage.feature import greycomatrix, greycoprops
 import numpy as np
 from skimage import data
 import rasterio

 path = 'C:\Users\GLCM_implementation\glasgow.tif'

 with rasterio.open(path, 'r') as src:
 import_file = src.read()
 img = import_file[0,:,:] #i need only the two dimentions (height, 
 width)
 print img.shape


 #calculate the GLCM specifying the distance, direction(4 directions) and 
 number of grey levels
 GLCM = greycomatrix(img, [1], [0, np.pi/4, np.pi/2, 3*np.pi/4],levels=8, 
 symmetric=False, normed=True)
 #list(GLCM[:,:,0,2])


 #Calculate texture statistics
 contrast = greycoprops(GLCM, 'contrast')

 

Re: GLCM calculation using scikit-learn. Error when using greycomatrix

[ioannisgkan259]

Hello Juan,

I want to thank you for your quick reply to my question the last time and i 
do not want to bother you with my questions. I always try to solve my 
problems before posting my question here. Sometime, i just cannot figure 
out how to fix the error.

What i want to do is to create a GLCM image using different  grey levels 
(8-bit, 16-bit and 32-bit) and compare the results. The type of the image i 
am using is a float32. I used the '' rescale intensity'' as you suggested, 
i converted the image to 8-bit and it worked. When i try to rescale the 
image in 16-bit and 32-bit for some reason it doesn't work.

i get the error:


AssertionErrorTraceback (most recent call 
last) in () 22  23 for j in 
xrange(img.shape[1] ):---> 24 glcm = greycomatrix(rescale, [1], [0],  
symmetric = True, normed = True ) 25  26 testraster1[i,j] = 
greycoprops(glcm, 'contrast')
C:\Anaconda2\lib\site-packages\skimage\feature\texture.pyc in 
greycomatrix(image, distances, angles, levels, symmetric, normed)101 
image = np.ascontiguousarray(image)102 assert image.min() >= 0--> 103   
  assert image.max() < levels104 image = image.astype(np.uint8)105  
   distances = np.ascontiguousarray(distances, dtype=np.float64)
AssertionError: 

It seems that i am allowed only to use an 8-bit image for the GLCM calculation.
So, does this code in scikit-image work only for 8-bit images? is there another 
way to solve my problem?
I will keep trying to find a solution but if you have an idea on how to solve 
this problem, please tell me

Thank you in advance
Ioannis



On Monday, 19 September 2016 15:41:07 UTC+1, ioannis...@gmail.com wrote:
>
> Thank you very much Juan for your quick reply.
> That was helpful :)
>
> Ioannis
>
> On Monday, 19 September 2016 01:03:45 UTC+1, Juan Nunez-Iglesias wrote:
>>
>> Hi Ioannis,
>>
>> Unfortunately the levels keyword is used as a hint to the function about 
>> the number of levels when the image is uint16, because the possible number 
>> of levels is huge. But if you want to convert the image to those levels, 
>> you have to do it manually. I suggest you look at the "rescale_intensity" 
>> function:
>>
>>
>> http://scikit-image.org/docs/dev/api/skimage.exposure.html#skimage.exposure.rescale_intensity
>>
>> and process your image before passing it to the glcm function.
>>
>> I hope this helps! Keep pinging if you have more questions. =)
>>
>> Juan.
>>
>> On Sun, Sep 18, 2016 at 4:57 AM,  wrote:
>>
>>> Hello everyone,
>>>
>>> I am using a SAR image (16-bit) and trying to implement GLCM algorithm 
>>> using sciki-learn. When trying to calculate the GLCM using greycomatrix i 
>>> get the following error: 
>>>
>>> assert image.max() < levels. It says that the maximum value of the image 
>>> intensity must be less than the number of grey levels.
>>> Because the SAR image is really big, i want to reduce the calculation time 
>>> by reducing the levels to 8. 
>>> Even if i remove the parameter 'level=8' when using greycomatrix, still 
>>> gives me the same error
>>>
>>> My code is the following:
>>>
>>> from skimage.feature import greycomatrix, greycoprops
>>> import numpy as np
>>> from skimage import data
>>> import rasterio
>>>
>>> path = 'C:\Users\GLCM_implementation\glasgow.tif'
>>>
>>> with rasterio.open(path, 'r') as src:
>>> import_file = src.read()
>>> img = import_file[0,:,:] #i need only the two dimentions (height, width)
>>> print img.shape
>>> 
>>>
>>> #calculate the GLCM specifying the distance, direction(4 directions) and 
>>> number of grey levels
>>> GLCM = greycomatrix(img, [1], [0, np.pi/4, np.pi/2, 3*np.pi/4],levels=8, 
>>> symmetric=False, normed=True)
>>> #list(GLCM[:,:,0,2])
>>>
>>>
>>> #Calculate texture statistics
>>> contrast = greycoprops(GLCM, 'contrast')
>>>
>>> dissimilarity = greycoprops(GLCM, 'dissimilarity')
>>>
>>> homogeneity = greycoprops(GLCM, 'homogeneity')
>>>
>>> energy = greycoprops(GLCM, 'energy')
>>>
>>> correlation = greycoprops(GLCM, 'correlation')
>>>
>>> ASM = greycoprops(GLCM, 'ASM')
>>>
>>>
>>>
>>> Error message:
>>>
>>>  101 image = np.ascontiguousarray(image)102 assert image.min() 
>>> >= 0--> 103 assert image.max() < levels104 image = 
>>> image.astype(np.uint8)105 distances = 
>>> np.ascontiguousarray(distances, dtype=np.float64)
>>> AssertionError: 
>>>
>>>
>>> I would appreciate any help.
>>> Thank you in advance
>>>
>>> Ioannis 
>>>
>>> -- 
>>> You received this message because you are subscribed to the Google 
>>> Groups "scikit-image" group.
>>> To unsubscribe from this group and stop receiving emails from it, send 
>>> an email to scikit-image...@googlegroups.com.
>>> To post to this group, send email to scikit...@googlegroups.com.
>>> To view this discussion on the web, visit 
>>> https://groups.google.com/d/msgid/scikit-image/520f5f2b-4750-4b56-a40b-28b938b750d8%40googlegroups.com
>>>  
>>>