-
Notifications
You must be signed in to change notification settings - Fork 1
/
tool.py
255 lines (228 loc) · 8.96 KB
/
tool.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
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
# -*- coding:utf-8 -*-
from time import time
from imutils.perspective import four_point_transform
from imutils import contours
import numpy as np
import concurrent.futures
import os
import cv2 as cv
from progress.bar import IncrementalBar
listeningCA = None
readingCA = None
fillingCA = None
workingPath = os.path.dirname(os.path.abspath(__file__))
with open(workingPath + "\\" + "answer.txt", 'r') as f:
content = f.readlines()
if len(content) < 3:
raise ValueError('File does not have enough lines')
listeningCALine = content[0].strip()
readingCALine = content[1].strip()
fillingCALine = content[2].strip()
listeningCA = [c for c in listeningCALine]
readingCA = [c for c in readingCALine]
fillingCA = [c for c in fillingCALine]
def showIMG(img):
t= str(time())
def click_event(event, x, y, flags, param):
# 如果鼠标左键被单击
if event == cv.EVENT_LBUTTONDOWN:
# 在点击位置绘制一个圆圈
cv.circle(img, (x, y), 3, (0, 0, 255), -1)
# 在点击位置显示坐标
text = f'({x}, {y})'
cv.putText(img, text, (x - 20 , y - 5), cv.FONT_HERSHEY_SIMPLEX, 0.8, (0, 0, 255), 2)
cv.imshow(t, img)
cv.namedWindow(t,cv.WINDOW_NORMAL)
cv.imshow(t, img)
cv.setMouseCallback(t, click_event)
def resize(img,h):
height, width = img.shape[:2]
aspect_ratio = h / height
new_width = int(width * aspect_ratio)
return cv.resize(img, (new_width, h))
def thresh(img,size):
'''对图像二值化'''
return cv.adaptiveThreshold(img, 255, cv.ADAPTIVE_THRESH_GAUSSIAN_C, cv.THRESH_BINARY_INV, size, 4)
def adjustBrightnessContrast(img, brightness=0, contrast=0):
# 调整亮度和对比度
b = brightness / 255.0
c = contrast / 255.0
k = np.tan((45 + 44 * c) / 180 * np.pi)
img = (img - 127.5 * (1 - b)) * k + 127.5 * (1 + b)
# 对图像进行裁剪，保证像素值在0-255之间
img = np.clip(img, 0, 255).astype(np.uint8)
return img
def ansList(imgName):
# 加载一个图片到opencv中
img = cv.imread(workingPath + "\\pic\\" + imgName)
img = resize(img,1280)
gray = cv.cvtColor(img, cv.COLOR_BGR2GRAY)
gray = cv.GaussianBlur(gray,(5,5),0)
# gray = adjustBrightnessContrast(gray,20,120)
# showIMG(gray)
thresh2 = thresh(gray,131)
# showIMG(thresh2)
# 寻找轮廓
cts, hierarchy = cv.findContours(thresh2.copy(), cv.RETR_EXTERNAL, cv.CHAIN_APPROX_SIMPLE)
# 给轮廓加标记，便于我们在原图里面观察，注意必须是原图才能画出红色，灰度图是没有颜色的
# cv.drawContours(img, cts, -1, (0, 0, 255), 7)
# showIMG(img)
# # 按面积大小对所有的轮廓排序
list = sorted(cts, key=cv.contourArea, reverse=True)
# cv.drawContours(img, list[0], -1, (0, 0, 255), 7)
# showIMG(img)
for c in list:
# 周长，第1个参数是轮廓，第二个参数代表是否是闭环的图形
peri = 0.02 * cv.arcLength(c, True)
# 获取多边形的所有定点，如果是四个定点，就代表是矩形
approx = cv.approxPolyDP(c, peri, True)
# 打印定点个数
# print("顶点个数：", len(approx))
if len(approx) == 4: # 矩形
# cv.drawContours(img, c, -1, (0, 0, 255), 7)
# showIMG(img)
# 透视变换提取原图内容部分
origin_sheet = four_point_transform(img, approx.reshape(4, 2))
# 透视变换提取灰度图内容部分
gray_sheet = four_point_transform(gray, approx.reshape(4, 2))
# showIMG(origin_sheet)
# cv.namedWindow('tx',cv.WINDOW_NORMAL)
# cv.imshow("tx", gray_sheet)
break
gray_sheet = resize(gray_sheet,1040)
origin_sheet = resize(origin_sheet,1040)
thresh2 = thresh(gray_sheet,101)
kernel = cv.getStructuringElement(cv.MORPH_RECT, (9, 9))
erosion = cv.erode(thresh2, kernel, iterations = 1)
# showIMG(erosion)
# 继续寻找轮廓
ans_cnt, ans_hierarchy = cv.findContours(erosion.copy(), cv.RETR_TREE, cv.CHAIN_APPROX_SIMPLE)
# print("找到轮廓个数：", len(ans_cnt))
# 使用红色标记所有的轮廓
# cv.drawContours(origin_sheet,ans_cnt,-1,(0,0,255),2)
listeningAnsCross=[]
readingAnsCross=[]
fillingAnsCross=[]
listeningAns=[]
readingAns=[]
fillingAns=[]
for cross in ans_cnt:
# 通过矩形，标记每一个指定的轮廓
x, y, w, h = cv.boundingRect(cross)
# ar = w / float(h)
if x>8 and w >= 30 and w <=80 and h >= 15 and h <= 60:
# and ar >= 0.5 and ar <= 3
# 标记
# cv.putText(origin_sheet, str(x)+ ',' + str(y), (x - 5 , y - 5), cv.FONT_HERSHEY_SIMPLEX, 0.8, (0, 0, 255), 2)
if y < 310:
cv.rectangle(origin_sheet, (x, y), (x + w, y + h), (0, 0, 255), 2)
listeningAnsCross.append(cross)
elif y<790:
cv.rectangle(origin_sheet, (x, y), (x + w, y + h), (0, 0, 255), 2)
readingAnsCross.append(cross)
else:
cv.rectangle(origin_sheet, (x, y), (x + w, y + h), (0, 0, 255), 2)
fillingAnsCross.append(cross)
# showIMG(origin_sheet)
listeningAnsCross = contours.sort_contours(listeningAnsCross)[0]
readingAnsCross = contours.sort_contours(readingAnsCross)[0]
fillingAnsCross = contours.sort_contours(fillingAnsCross)[0]
for i in listeningAnsCross:
x, y, w, h = cv.boundingRect(i)
if y<110:
listeningAns.append('A')
elif y<169:
listeningAns.append('B')
else:
listeningAns.append("C")
for i in readingAnsCross:
x, y, w, h = cv.boundingRect(i)
if y<390:
readingAns.append('A')
elif y<450:
readingAns.append('B')
elif y<505:
readingAns.append("C")
elif y<560:
readingAns.append("D")
elif y<610:
readingAns.append("E")
elif y<670:
readingAns.append("F")
else:
readingAns.append("G")
for i in fillingAnsCross:
x, y, w, h = cv.boundingRect(i)
if y<850:
fillingAns.append('A')
elif y<910:
fillingAns.append('B')
elif y<965:
fillingAns.append("C")
else:
fillingAns.append("D")
return ([listeningAns,readingAns,fillingAns],origin_sheet)
def worker(imgName):
finalScore = 0
ori_sheet = None
ans = None
fail = 0
# 0 success, 1 length fail, 2 other fail
try:
ans,ori_sheet = ansList(imgName)
fail = 1 if len(ans[0])!=len(listeningCA) or len(ans[1])!=len(readingCA) or len(ans[2])!=len(fillingCA) else 0
except Exception as e:
fail = 2
for CANum, CA in enumerate(listeningCA):
try:
if ans[0][CANum] == CA:
finalScore += 1.5
except Exception as e:
fail = 2
for CANum, CA in enumerate(readingCA):
try:
if ans[0][CANum] == CA:
finalScore += 2.5
except Exception as e:
fail = 2
for CANum, CA in enumerate(fillingCA):
try:
if ans[0][CANum] == CA:
finalScore += 1
except Exception as e:
fail = 2
if fail!=0:
code=['success','length fail','other fail']
with open(workingPath+ "\\result\\" + 'wrong.txt', "w") as file:
file.write(imgName + '/' + code[fail] +'\n')
if ans!=None:
# 设置文本信息
text = str(finalScore)
font = cv.FONT_HERSHEY_SIMPLEX
font_scale = 4
thickness = 2
color = (0, 0, 255)
# 获取文本大小
text_size, _ = cv.getTextSize(text, font, font_scale, thickness)
# 计算文本位置
text_x = ori_sheet.shape[1] - text_size[0] - 10
text_y = ori_sheet.shape[0] - text_size[1] - 10
# 在图像上添加文本
cv.putText(ori_sheet, text, (text_x, text_y), font, font_scale, color, thickness, cv.LINE_AA)
cv.imwrite(workingPath+ "\\result\\" + imgName, ori_sheet)
img_files = [] # 保存 JPG 文件名的数组
# 遍历目录下的所有文件
for filename in os.listdir(workingPath + '\\pic'):
# 判断是否为 JPG 文件
if filename.lower().endswith('.jpg'):
# 将文件名保存到数组中
img_files.append(filename)
# ansList('1.jpg')
futures = []
with concurrent.futures.ThreadPoolExecutor(max_workers=10) as executor:
for i in img_files:
futures.append(executor.submit(worker, i))
with IncrementalBar('Processing', max=len(futures)) as bar:
for future in concurrent.futures.as_completed(futures):
bar.next()
# cv.waitKey(0)