-
Notifications
You must be signed in to change notification settings - Fork 0
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
João Mamede
committed
Sep 28, 2023
1 parent
b500aba
commit 7728c22
Showing
2 changed files
with
277 additions
and
12 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,237 @@ | ||
| import scipy.ndimage as ndi | ||
| #import cv2 | ||
| import numpy as np | ||
| #import matplotlib.pyplot as plt | ||
| #import pims | ||
| #import libtiff | ||
| #import time | ||
| #import csv | ||
| _8bit = float(2**8-1) | ||
| _16bit = float(2**16-1) | ||
| ratio = _8bit /_16bit | ||
|
|
||
| def points_in_mask(coords,mask): | ||
| ''' | ||
| Input: | ||
| coords: list of coordinates (numpy) | ||
| mask: masked image with numbered labels (numpy) | ||
| Return: | ||
| values_at_coords: with boolean array reporting which mask does the coordinate belongs to | ||
| ''' | ||
| import numpy as np | ||
| # coords_int = np.round(coords).astype(int) # or np.floor, depends | ||
| coords_int = np.floor(coords).astype(int) # or np.floor, depends | ||
| try: | ||
| values_at_coords = mask[tuple(coords_int.T)] | ||
| except: values_at_coords = [] | ||
| # .astype(np.int) | ||
| # values_at_coords = mask[tuple(coords_int)].astype(np.int) | ||
| print(values_at_coords) | ||
| return values_at_coords | ||
|
|
||
|
|
||
| def multiply(real_events, image): | ||
| import numpy as np | ||
| image = real_events*image | ||
| return image | ||
|
|
||
| ##keep for now for the #make masks section | ||
| def make_labels_trackpy(image,mass,size=9,_separation=3,_numba=True,max_mass=0,_round=True): | ||
| import trackpy as tp | ||
| import scipy.ndimage as ndi | ||
| from scipy.ndimage.morphology import binary_dilation | ||
|
|
||
| if image.ndim == 2: | ||
| _size = size | ||
| elif image.ndim == 3: | ||
| _size = (3, size, size) | ||
| # _size = (9, size, size) | ||
| # ~ dotrack(ficheiro, plotgraph=True,_numba=True,massa=1500,tamanho=13,dist=250,memoria=1,stub=3,frame=19,colourIDX=0): | ||
| if image.ndim == 2: | ||
| if _numba: | ||
| f = tp.locate(image,diameter=size,separation=_separation,minmass=mass,engine='numba') | ||
| else: | ||
| f = tp.locate(image,diameter=size,separation=_separation, minmass=mass) | ||
| elif image.ndim == 3: | ||
| if _numba: | ||
| f = tp.locate(image,diameter=_size,separation = (3, 3, 3), | ||
| minmass=mass,engine='numba') | ||
| else: | ||
| f = tp.locate(image,diameter=_size,separation = (3, 3, 3), | ||
| minmass=mass) | ||
| # size = (11, 13, 13) | ||
|
|
||
| if max_mass > 0: | ||
| f = f[f['mass'] <= max_mass] | ||
| #outputsomehow is 3D, we want 2 | ||
| if image.ndim == 2: | ||
| coords = np.dstack((round(f.y),round(f.x)))[0].astype(int) | ||
| elif image.ndim == 3: | ||
| coords = np.dstack((round(f.z),round(f.y),round(f.x)))[0].astype(int) | ||
|
|
||
|
|
||
|
|
||
| #this is super slow | ||
| # ~ masks = tp.masks.mask_image(coords,np.ones(image.shape),size/2) | ||
|
|
||
| #This is faster | ||
| if image.ndim == 2: | ||
| r = (size-1)/2 # Radius of circles | ||
| #make 3D compat | ||
| disk_mask = tp.masks.binary_mask(r,image.ndim) | ||
| # Initialize output array and set the maskcenters as 1s | ||
| out = np.zeros(image.shape,dtype=bool) | ||
| #check if there's a problem with subpixel masking | ||
| out[coords[:,0],coords[:,1]] = 1 | ||
| # Use binary dilation to get the desired output | ||
| out = binary_dilation(out,disk_mask) | ||
| labels, nb = ndi.label(out) | ||
|
|
||
| if _round: | ||
| return labels, coords | ||
| else: | ||
| if image.ndim == 2: | ||
| coords = np.dstack((f.y,f.x))[0] | ||
| return labels, coords | ||
| elif image.ndim == 3: | ||
| coords = np.dstack((f.z,f.y,f.x))[0] | ||
| return None, coords | ||
|
|
||
|
|
||
|
|
||
| def make_labels_trackpy_links(image,j,size=5): | ||
| import trackpy as tp | ||
| import scipy.ndimage as ndi | ||
| from scipy.ndimage.morphology import binary_dilation | ||
|
|
||
|
|
||
| #outputsomehow is 3D, we want 2 | ||
| coords = np.dstack((round(j.y),round(j.x)))[0].astype(int) | ||
|
|
||
| #this is super slow | ||
| # ~ masks = tp.masks.mask_image(coords,np.ones(image.shape),size/2) | ||
|
|
||
| #This is faster | ||
| r = (size-1)/2 # Radius of circles | ||
| #make 3D compat | ||
| disk_mask = tp.masks.binary_mask(r,image.ndim) | ||
| # Initialize output array and set the maskcenters as 1s | ||
| out = np.zeros(image.shape,dtype=bool) | ||
| #check if there's a problem with subpixel masking | ||
| out[coords[:,0],coords[:,1]] = 1 | ||
| # Use binary dilation to get the desired output | ||
| out = binary_dilation(out,disk_mask) | ||
|
|
||
|
|
||
| labels, nb = ndi.label(out) | ||
|
|
||
| return labels, coords | ||
|
|
||
| def simple_labels(image): | ||
| labels, nb = ndi.label(image) | ||
| return labels | ||
|
|
||
| def mean_calc(values,value_max): | ||
| import numpy as np | ||
| #Only the ones above threshold will be averaged don't worry about | ||
| #the b>value_max because the threshold is the same. | ||
| a = np.asarray([value['MeanIntensity'] for value in values]) | ||
| b = np.asarray([value['MaxIntensity'] for value in values]) | ||
| return np.mean(a[b > value_max]) | ||
|
|
||
| def mean_calc_total(values): | ||
| import numpy as np | ||
| return np.mean([value['MeanIntensity'] for value in values]) | ||
|
|
||
| def max_calc_total(values): | ||
| import numpy as np | ||
| return np.mean([value['MaxIntensity'] for value in values]) | ||
|
|
||
| def listify(data,datatype): | ||
| if datatype == 'max': datatype = 'MaxIntensity' | ||
| elif datatype == 'mean': datatype = 'MeanIntensity' | ||
| return np.asarray([value[datatype] for value in data]) | ||
|
|
||
| def rebin(arr, new_shape): | ||
| from PIL import Image | ||
| return np.array(Image.fromarray(arr).resize(new_shape,resample=Image.NEAREST)) | ||
|
|
||
| #~ @v.parallel(block=True) | ||
| def othercolor(colour,labels): | ||
| import numpy as np | ||
| import scipy.ndimage as ndi | ||
| #This gives an array with max mean and stdev values per each label | ||
|
|
||
| data = {} | ||
| data['max'] = ndi.maximum(colour,labels,np.arange(1,labels.max()+1)) | ||
| data['mean'] = ndi.mean(colour,labels,np.arange(1,labels.max()+1)) | ||
| data['stdev'] = ndi.standard_deviation(colour,labels,np.arange(1,labels.max()+1)) | ||
| data['median'] = ndi.median(colour,labels,np.arange(1,labels.max()+1)) | ||
| try: | ||
| temp = np.array(ndi.measurements.center_of_mass( | ||
| colour,labels,np.arange(1,labels.max()+1))) | ||
| # ~ print(temp[:,0]) | ||
| data['COMY'] = temp[:,0] | ||
| data['COMX'] = temp[:,1] | ||
| except: 'Something went wrong with centroids' | ||
| return data | ||
|
|
||
|
|
||
| def othercolor2(colour,labels,treshold): | ||
| import numpy as np | ||
| import scipy.ndimage as ndi | ||
| '''Output: A dict() with max mean and stdev values per each label''' | ||
|
|
||
| data = {} | ||
| data['max'] = ndi.maximum(colour,labels,np.arange(1,labels.max()+1)) | ||
| data['mean'] = ndi.mean(colour,labels,np.arange(1,labels.max()+1)) | ||
| data['stdev'] = ndi.standard_deviation(colour,labels,np.arange(1,labels.max()+1)) | ||
| data['median'] = ndi.median(colour,labels,np.arange(1,labels.max()+1)) | ||
| data['mean_pos'] = data['mean'][data['max'] > treshold] | ||
| data['median_pos'] = data['median'][data['max'] > treshold] | ||
| data['stdev_pos'] = data['stdev'][data['max'] > treshold] | ||
| data['max_pos'] = data['max'][data['max'] > treshold] | ||
|
|
||
| return data | ||
|
|
||
| def contrast_img(img,min_,max_ ): | ||
| img[img>max_]=max_ | ||
| img[img<min_]=min_ | ||
| img -= min_ | ||
| img = img * (_16bit/float(max_-min_)) | ||
| return img | ||
|
|
||
|
|
||
| def convert16to8bits_gpu(x,display_min=0,display_max=2**16-1): | ||
| import cupy as cp | ||
| def display(image, display_min, display_max): # copied from Bi Rico | ||
| # Here I set copy=True in order to ensure the original image is not | ||
| # modified. If you don't mind modifying the original image, you can | ||
| # set copy=False or skip this step. | ||
| # image = cp.array(image, copy=FalseTrue) | ||
| image.clip(display_min, display_max, out=image) | ||
| image -= display_min | ||
| cp.floor_divide(image, (display_max - display_min + 1) / 256, | ||
| out=image, casting='unsafe') | ||
| return image.astype(cp.uint8) | ||
|
|
||
| lut = cp.arange(2**16, dtype='uint16') | ||
| lut = display(lut, display_min, display_max) | ||
| return cp.asnumpy(cp.take(lut, x)) | ||
|
|
||
| def convert16to8bits(x,display_min=0,display_max=2**16-1): | ||
| def display(image, display_min, display_max): # copied from Bi Rico | ||
| # Here I set copy=True in order to ensure the original image is not | ||
| # modified. If you don't mind modifying the original image, you can | ||
| # set copy=False or skip this step. | ||
| # image = cp.array(image, copy=FalseTrue) | ||
| image.clip(display_min, display_max, out=image) | ||
| image -= display_min | ||
| np.floor_divide(image, (display_max - display_min + 1) / 256, | ||
| out=image, casting='unsafe') | ||
| return image.astype(np.uint8) | ||
|
|
||
| lut = np.arange(2**16, dtype='uint16') | ||
| lut = display(lut, display_min, display_max) | ||
| return np.take(lut, x) |