-
Notifications
You must be signed in to change notification settings - Fork 4
/
img_transform.py
123 lines (113 loc) · 4.54 KB
/
img_transform.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
#!/usr/bin/env python
# encoding: utf-8
"""
@version: 0.1
@author: springhser
@license: Apache Licence
@contact: endoffight@gmail.com
@site: http://www.springhser.com
@software: PyCharm Community Edition
@file: img_transform.py
@time: 2017/12/30 17:58
"""
import numpy as np
import cv2
from matplotlib import pyplot as plt
# 线段类
class Line:
def __init__(self, line):
# x1, y1, x2, y2 = l # 前两个数为起点，后两个数为终点
self.x1 = line[0]
self.y1 = line[1]
self.x2 = line[2]
self.y2 = line[3]
# 线段中点的坐标, 为排序使用
self.half_x = (self.x1 + self.x2) / 2
self.half_y = (self.y1 + self.y2) / 2
# 求出与另外一条不平行线段延长线的交点
def get_cross_point(self, l_a):
a1 = self.y2 - self.y1
b1 = self.x1 - self.x2
c1 = a1 * self.x1 + b1 * self.y1
a2 = l_a.y2 - l_a.y1
b2 = l_a.x1 - l_a.x2
c2 = a2 * l_a.x1 + b2 * l_a.x2
d = a1 * b2 - a2 * b1
if d is 0: # 平行或共线的情况
raise ValueError
return (1. * (b2 * c1 - b1 * c2) / d, 1. * (a1 * c2 - a2 * c1) / d)
# 对图像进行预处理求出图像中的直线
def img_process(old_img):
# 重新设定图像大小，方便计算
height = old_img.shape[0]
weight = old_img.shape[1]
min_weight = 200
scale = min(10., weight * 1. / min_weight)
new_h = int(height * 1. / scale)
new_w = int(weight * 1. / scale)
new_img = cv2.resize(old_img, (new_w, new_h))
gray_img = cv2.cvtColor(new_img, cv2.COLOR_BGR2GRAY) # 转为灰度图像
# 利用Canny 边缘检测和霍夫变换提取直线
highthreshold = cv2.threshold(gray_img, 0, 255, cv2.THRESH_BINARY | cv2.THRESH_OTSU)[0] # http://blog.csdn.net/on2way/article/details/46812121
lowthreshold = highthreshold * 0.2
canny_img = cv2.Canny(gray_img, lowthreshold, highthreshold)
return cv2.HoughLinesP(canny_img, 1, np.pi / 180, new_w // 3, 50, new_w // 3, 20), new_img
# 获得四个目标点
def get_target_points(lines, img):
# 分别提取出水平和垂直的线段
# print(lines)
lines_h = [] # 存放接近水平的线段
lines_v = [] # 存放接近垂直的线段
lines1 = lines[:, 0, :] # 提取为二维
for l in lines1:
# print(l)
line = Line(l)
if abs(line.x1 - line.x2) > abs(line.y1 - line.y2):
lines_h.append(line)
else:
lines_v.append(line)
# 如果线段数不够两条, 直接用原图像的边缘替代
if len(lines_h) <= 1:
if not lines_h or lines_h[0].half_y > img.shape[0]/2:
lines_h.append(Line((0, 0, img.shape[1] - 1, 0)))
if not lines_h or lines_h[0].half.y <= img.shape[0] / 2:
lines_h.append(Line((0, img.shape[0] - 1, img.shape[1] - 1, img.shape[0] - 1)))
if len(lines_v) <= 1:
if not lines_v or lines_v[0].half_x > img.shape[1] / 2:
lines_v.append(Line((0, 0, 0, img.shape[0] - 1)))
if not lines_v or lines_v[0].c_x <= img.shape[1] / 2:
lines_v.append(Line((img.shape[1] - 1, 0, img.shape[1] - 1, img.shape[0] - 1)))
# 获取最靠近边缘的四条线段求出他们的交点
lines_h.sort(key=lambda line: line.half_y)
lines_v.sort(key=lambda line: line.half_x)
return [lines_h[0].get_cross_point(lines_v[0]),
lines_h[0].get_cross_point(lines_v[-1]),
lines_h[-1].get_cross_point(lines_v[0]),
lines_h[-1].get_cross_point(lines_v[-1])]
# 做透视变换
def per_transform(target_points, old_img):
height = old_img.shape[0]
weight = old_img.shape[1]
min_weight = 200
scale = min(10., weight * 1. / min_weight)
# 恢复为原图像大小
for i, p in enumerate(target_points):
x, y = p
target_points[i] = (x * scale, y * scale)
# 原图像的四个点
four_points= np.array(((0, 0),
(weight - 1, 0),
(0, height - 1),
(weight - 1, height - 1)),
np.float32)
target_points = np.array(target_points, np.float32)
M = cv2.getPerspectiveTransform(target_points, four_points)
return cv2.warpPerspective(old_img, M, (weight, height))
if __name__ == '__main__':
old_img = cv2.imread("001.jpg")
lines, new_img = img_process(old_img)
t_points = get_target_points(lines, new_img)
revert_img = per_transform(t_points, old_img)
plt.imshow(revert_img,)
plt.xticks([]), plt.yticks([]) # to hide tick values on X and Y axis
plt.show()