codebase / stimlib

Commit c8b1298401f3a2388c16aa30a0820d4f39f5522d

Authored by sberisha
1 parent eccf10ff

renamed spectral.py to stim_spectral.py; modified envi_batch_predict to return R… 

…GB array in case user wants to save the fig
Showing 2 changed files with 5 additions and 56 deletions   Show diff stats
Inline Side-by-side
python/classify.py
  Show/Hide comments   View file @c8b1298
... ... @@ -12,7 +12,7 @@ import sklearn.metrics
12 12 import scipy
13 13 import scipy.misc
14 14 import envi
15   -import spectral
  15 +import stim_spectral
16 16 import random
17 17 import progressbar
18 18 import matplotlib.pyplot as plt
... ... @@ -176,11 +176,13 @@ def envi_batch_predict(E, C, batch=10000):
176 176 else:
177 177 Tv = numpy.concatenate((Tv, C.predict(Fv.transpose()).transpose()), 0)
178 178 tempmask = E.batchmask()
179   - Lv = spectral.unsift2(Tv, tempmask)
  179 + Lv = stim_spectral.unsift2(Tv, tempmask)
180 180 Cv = label2class(Lv.squeeze(), background=0)
181 181 RGB = class2color(Cv)
182 182 plt.imshow(RGB)
183 183 plt.pause(0.05)
184 184 Fv = E.loadbatch(batch)
185 185 i = i + 1
186   - bar.update(len(Tv))
187 186 \ No newline at end of file
  187 + bar.update(len(Tv))
  188 +
  189 + return RGB
188 190 \ No newline at end of file
... ...
python/spectral.py deleted
View file @eccf10f
1   -# -*- coding: utf-8 -*-
2   -"""
3   -Created on Sun Jul 23 13:52:22 2017
4   -
5   -@author: david
6   -"""
7   -import numpy
8   -
9   -#sift a 2D hyperspectral image into a PxB matrix where P is the number of pixels and B is the number of bands
10   -def sift2(I, mask = []):
11   -
12   - #get the shape of the input array
13   - S = I.shape
14   -
15   - #convert that array into a 1D matrix
16   - M = numpy.reshape(I, (S[0], S[1] * S[2]))
17   -
18   - #gif no mask is provided, just return all pixels
19   - if mask == []:
20   - return M
21   -
22   - #if a mask is provided, only return pixels corresponding to that mask
23   - flatmask = numpy.reshape(mask, (S[1] * S[2]))
24   - i = numpy.flatnonzero(flatmask) #get the nonzero indices
25   - return M[:, i] #return pixels corresponding to the masked values
26   -
27   -def unsift2(M, mask):
28   -
29   - #get the size of the input matrix
30   - S = M.shape
31   -
32   - #count the number of nonzero values in the mask
33   - nnz = numpy.count_nonzero(mask)
34   -
35   - #the number of masked values should be the same as the number of pixels in the input matrix
36   - if len(S) == 1:
37   - if not S[0] == nnz:
38   - print("ERROR: expected " + str(nnz) + " pixels based on the mask but there are " + str(S[0]) + " in the matrix.")
39   - elif not S[1] == nnz:
40   - print("ERROR: expected " + str(nnz) + " pixels based on the mask but there are " + str(S[1]) + " in the matrix.")
41   -
42   -
43   - i = numpy.nonzero(mask)
44   -
45   - if len(S) == 1:
46   - I = numpy.zeros((1, mask.shape[0], mask.shape[1]), dtype=M.dtype)
47   - else:
48   - I = numpy.zeros((M.shape[0], mask.shape[0], mask.shape[1]), dtype=M.dtype)
49   - I[:, i[0], i[1]] = M
50   - return I
51   -
52   -#create a function that sifts a color image
53   -#input: image name, mask
54 0 \ No newline at end of file