-
Notifications
You must be signed in to change notification settings - Fork 0
/
hough.py
156 lines (129 loc) · 5.13 KB
/
hough.py
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
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
import cv2
import numpy as np
import imutils
import pandas as pd
RESIZE_WIDTH = 500
SAME_ROW_Y_AXIS_OFFSET = 15
def extract_ROI_value(data, MEAN_R, img6, circle_mask):
#TODO : exception
answer_circle_ROI = img6[data[1]-MEAN_R:data[1]+MEAN_R,
data[0]-MEAN_R:data[0]+MEAN_R,
:].copy()
answer_circle_ROI = cv2.bitwise_and(answer_circle_ROI, answer_circle_ROI, mask=circle_mask)
return np.sum(answer_circle_ROI)
def min_ROI_data(dataset):
min_ROI = dataset[0]['roi']
answer_data = dataset[0]['data']
answer_option = 1
for option, data in enumerate(dataset):
if data['roi'] < min_ROI:
min_ROI = data['roi']
answer_data = data['data']
answer_option = option + 1
return (answer_option, answer_data)
def group_circle_by_y(df, option_number, r, img6, circle_mask):
# collect circle data
# according to 'y', grouping circle
group = dict()
grouped = False
for index, data in df.iterrows():
for key in group.keys():
int_key = int(key)
if int_key - SAME_ROW_Y_AXIS_OFFSET < data['y'] and data['y'] < int_key + SAME_ROW_Y_AXIS_OFFSET:
group[key].append(dict(data=data, roi=extract_ROI_value(data, r, img6, circle_mask)))
grouped = True
break
if not grouped:
group[str(data['y'])] = [dict(data=data, roi=extract_ROI_value(data, r, img6, circle_mask))]
grouped = False
keys = list(group.keys())
for key in keys:
# check error detect
if len(group[key]) <= 1:
group.pop(key)
continue
#TODO
#if len(group[key]) > option_number:
# return []
group[key] = sorted(group[key], key=lambda x: x['data']['x'])
return group
def collect_answer(group, question_number):
# min_ROI_data would return a tuple! (option_order, data)
result = [min_ROI_data(group[key]) for key in group.keys()]
result = sorted(result, key=lambda x: x[1]['y'])
#for i in result:
# print('*'*50)
# print(i)
result = [choose[0] for choose in result]
#TODO : comment
#if len(result) != question_number:
# return []
return result
def show_image(img6, origin_img3):
cv2.imshow('detected circles',img6)
cv2.imshow('original pic', origin_img3)
cv2.waitKey(0)
cv2.destroyAllWindows()
def test_para(filename, para1, para2):
question_number = 5
option_number = 5
#filename = input('name : ')
#origin_img3 = cv2.imread(filename + '.jpeg')
origin_img3 = cv2.imread(filename)
origin_img3 = imutils.resize(origin_img3, width=RESIZE_WIDTH)
#img3 = cv2.GaussianBlur(origin_img3,(3, 3), 1)
img3 = cv2.medianBlur(origin_img3, 3)
img3 = cv2.GaussianBlur(img3,(3, 3), 1)
#img3 = cv2.blur(origin_img3, (3,3))
#img3 = origin_img3
#cv2.imshow('blur', img3)
img1 = cv2.cvtColor(img3,cv2.COLOR_BGR2GRAY)
circles = cv2.HoughCircles(img1,cv2.HOUGH_GRADIENT,1,20,param1=para1,param2=para2,minRadius=6,maxRadius=16)
circles = np.uint16(np.around(circles))
MEAN_R = np.uint16(np.ceil(np.mean(circles, axis=1)[0][2]))
img6 = origin_img3.copy()
circle_mask = np.zeros((MEAN_R*2, MEAN_R*2), dtype=np.uint8)
cv2.circle(img=circle_mask,
center=(MEAN_R, MEAN_R),
radius=MEAN_R,
color=(255, 255, 255),
thickness=cv2.FILLED
)
EROSION_KERNEL = np.ones((2,2), np.uint8)
for i in circles[0, :]:
answer_circle_ROI = img6[i[1]-MEAN_R:i[1]+MEAN_R,
i[0]-MEAN_R:i[0]+MEAN_R,
:]
answer_circle_ROI = cv2.bitwise_and(answer_circle_ROI, answer_circle_ROI, mask=circle_mask)
answer_circle_ROI = cv2.adaptiveThreshold(cv2.cvtColor(answer_circle_ROI, cv2.COLOR_BGR2GRAY),255,cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY,11,1)
answer_circle_ROI = cv2.erode(answer_circle_ROI, EROSION_KERNEL, iterations=1)
img6[i[1]-MEAN_R:i[1]+MEAN_R, i[0]-MEAN_R:i[0]+MEAN_R, :] = cv2.cvtColor(answer_circle_ROI, cv2.COLOR_GRAY2BGR)
#cv2.circle(img6,(i[0],i[1]),MEAN_R,(0,255,0),2)
#cv2.circle(img6,(i[0],i[1]),2,(0,0,255),1)
dataframe = pd.DataFrame(circles.reshape(-1, 3), columns=['x', 'y', 'r'])
group = group_circle_by_y(dataframe, option_number,MEAN_R, img6, circle_mask)
res = collect_answer(group, question_number)
return res
# 30 25
ind = 0
ans = [2, 3, 1 , 4, 5, 2, 5, 1]
all_file = ['d/1.jpeg', 'd/2.jpeg', 'd/3.jpeg', 'd/4.jpeg', 'd/5.jpeg', 'd/6.jpeg', 'd/7.jpeg', 'd/8.jpeg']
args1 = [98, 182, 266, 350]
args2 = [18]
for arg1 in args1:#100~350/18
for arg2 in args2:
count = 0
print('*'*50)
print(arg1, arg2)
for filename in all_file:
try:
result = test_para(filename, arg1, arg2)
except:
continue
if result == ans:
print(filename.split('/')[1].split('.')[0], end=" ")
count = count + 1
ind = ind + 1
print()
if count >= 6:
print(arg1, arg2)