DigitRecognitionPython/FindDigits.py at master · sn2234/DigitRecognitionPython · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67

import sys
import numpy as np
import matplotlib.pyplot as plt
from skimage import io

import Train
import SimpleNN2

# Load image as grayscale
fileName = r"..\iphonepuzzle_a-600x893.jpg"

img = io.imread(fileName, True)

# Load neural network
(nn, th1, th2) = SimpleNN2.loadNetwork(r"..\networkParams.bin")

# Try 28*28 slices

sliceSizeX = 28
sliceSizeY = 28

xSlices = img.shape[1] // sliceSizeX
ySlices = img.shape[0] // sliceSizeY

imgMod = img.copy()

for x in range(xSlices):
    for y in range(ySlices):
        xFr = x*sliceSizeX
        xTo = xFr + sliceSizeX

        yFr = y*sliceSizeY
        yTo = yFr + sliceSizeY

        subImage = img[yFr:yTo, xFr:xTo]

        #prob = SimpleNN2.predictProbability(nn, th1, th2, subImage.flatten()).flatten()
        cl = SimpleNN2.predictProbability(nn, th1, th2, subImage.flatten()).flatten()

        if cl[3] > 0.25 :
            print("Found at ({0},{1})".format(y, x))
            imgMod[yFr:yTo, xFr:xTo] *= 0.5
            #plt.imshow(subImage.reshape((28,28)))
            #break

plt.imshow(img, cmap=cm.gray)
plt.imshow(imgMod, cmap=cm.gray)
#for x in range(xSlices):
#    for y in range(ySlices):
#        xFr = x*sliceSizeX
#        xTo = xFr + sliceSizeX

#        yFr = y*sliceSizeY
#        yTo = yFr + sliceSizeY

#        subImage = img[yFr:yTo, xFr:xTo]

#        ax = plt.subplot(ySlices, xSlices, y*xSlices+x+1)
#        plt.set_cmap('gray')
#        plt.axis('off')
#        ax.imshow(subImage.reshape((28,28)))

#plt.subplots_adjust(hspace=-0.85)
#plt.gcf().set_size_inches(9, 9)

#plt.show()