-
Notifications
You must be signed in to change notification settings - Fork 0
/
util.py
88 lines (79 loc) · 2.91 KB
/
util.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
#!/usr/bin/env python
# -*- coding: utf-8 -*-
# @Time : 2017/12/20 09:18
# @Author : Guo Ziyao
import cv2
import numpy as np
from PIL import Image
import pytesseract
def show_img(img):
"""
显示图像,按任意键退出
"""
cv2.imshow('test', img)
cv2.waitKey(0)
def correct_angle(img):
"""
图像倾斜矫正。找到外接矩形,然后取两条宽的中点算斜率,求倾斜角度
:img: 二值化图像
:return: 返回图像对象
"""
blur = cv2.GaussianBlur(img, (51, 51), 0)
ret, binary_img = cv2.threshold(blur, 0, 255, cv2.THRESH_OTSU | cv2.THRESH_BINARY_INV)
edges = cv2.Canny(binary_img, 100, 200)
image, contours, hierarchy = cv2.findContours(edges, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
index = 0
for each in contours:
point = cv2.minAreaRect(each)[1]
if point[1] != 0 and point[0] != 0:
break
index += 1
rect = cv2.minAreaRect(contours[index])
# angle = rect[2]
# print angle
points = cv2.boxPoints(rect)
left, right, up, down = (1000, 0), (-1000, 0), (0, 1000), (0, -1000)
for each in points:
x, y = each
if y <= up[1]:
up = (x, y)
if y >= down[1]:
down = (x, y)
if x <= left[0]:
left = (x, y)
if x >= right[0]:
right = (x, y)
# 如果本身水平
if up[0] == down[0]:
return img
# print up, down, left, right
# cv2.rectangle(img, up, down, 1)
# show_img(img)
if up[0] > down[0]:
x1, y1 = (up[0] + right[0]) / 2.0, (up[1] + right[1]) / 2.0
x2, y2 = (down[0] + left[0]) / 2.0, (down[1] + left[1]) / 2.0
angle = np.arctan((y1 - y2) / (x1 - x2)) * 180 / np.pi
else:
x1, y1 = (up[0] + left[0]) / 2.0, (up[1] + left[1]) / 2.0
x2, y2 = (down[0] + right[0]) / 2.0, (down[1] + right[1]) / 2.0
angle = np.arctan((y1 - y2) / (x1 - x2)) * 180 / np.pi
rows, cols = img.shape[: 2]
# print angle
M = cv2.getRotationMatrix2D((int(rows / 2), int(cols / 2)), angle, 1)
img = cv2.warpAffine(img, M, (rows, cols), flags=cv2.INTER_CUBIC, borderMode=cv2.BORDER_REPLICATE)
# cv2.drawContours(img, contours, 0, (0, 0, 255), 1)
return img
def recgnize(img, lang):
cv2.imwrite('tmp.jpg', img)
# pytesseract 0.1.8
# http://hellosure.github.io/ocr/2014/10/11/tesseract-ocr
# -psm 7: only recognize one line
code = pytesseract.image_to_string(cv2.imread('tmp.jpg'), config="-psm 7 -c tessedit_char_whitelist='abcdefghijklmnopqrstuvwxyz' -l %s" % lang)
return code
if __name__ == '__main__':
img = cv2.imread('test/im0.png')
h, w = img.shape[: 2]
img = cv2.resize(img, None, fx=256.0 / w, fy=256.0 / w, interpolation=cv2.INTER_CUBIC)
gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
ret, binary_img = cv2.threshold(gray_img, 0, 255, cv2.THRESH_OTSU | cv2.THRESH_BINARY_INV)
show_img(correct_angle(binary_img))