-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathPlateTest.py
124 lines (103 loc) · 3.86 KB
/
PlateTest.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
import cv2
import numpy as np
# 车牌宽度按7.5的比例来分
img = cv2.imread('number_plate1.jpg')
# hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
# lower_blue = np.array([105, 43, 46])
# upper_blue = np.array([130, 255, 255])
# mask = cv2.inRange(hsv, lower_blue, upper_blue)
# cv2.imshow('a', mask)
plate = cv2.resize(img, None, fx=2, fy=2) # 放大两倍
copyplate = plate.copy()
gray = cv2.cvtColor(plate, cv2.COLOR_BGR2GRAY)
# eqImg = cv2.equalizeHist(gray)
# gaussian = cv2.GaussianBlur(eqImg, (3, 3), 0, 0, cv2.BORDER_DEFAULT)
# 中值滤波
# median = cv2.medianBlur(gaussian, 5)
# sobel = cv2.Sobel(median, cv2.CV_8U, 1, 0, ksize=3)
ret, binary = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
cv2.imshow('a', binary)
numyPic = np.array(binary)
row, colum = numyPic.shape
print(row, colum)
# 处理行白点
for x in range(row):
whiteNum = 0
white = 255
black = 0
flag = 0
up_down_num = 0
for y in range(colum):
# 行白点的数目大于一定值，判断这一行为直线白点,每个字符宽度小于1/6
if numyPic[x][y] > 125:
whiteNum += 1
# 判断跳变次数
if numyPic[x][y] != flag:
up_down_num += 1
flag = numyPic[x][y]
if whiteNum > colum * 0.55:
for i in range(colum):
numyPic[x][i] = 0
# 黑白跳变小于14次，就说明这行没有字符
if up_down_num < 14:
for i in range(colum):
numyPic[x][i] = 0
# # 扩大图像 是 （列行）
# test = cv2.resize(numyPic, (200,44),interpolation=cv2.INTER_CUBIC)
# returnimg, contours, hierarchy = cv2.findContours(numyPic, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
# cv2.drawContours(img, contours, 0, (0, 255, 0), -1)
cv2.imshow('aaa', numyPic)
edge = cv2.Canny(numyPic, 100, 200, True)
# returnImg, contours, hierarchy = cv2.findContours(edge, cv2.RETR_TREE, cv2.CHAIN_APPROX_SIMPLE)
#
# imgg = cv2.drawContours(copyplate, contours, -1, (0,0,255), 1)
#
# for i in range(len(contours)):
# cnt = contours[i]
# hull = cv2.convexHull(cnt)
# area = cv2.minAreaRect(cnt)
# print("第",i,"个面积：",area)
def scanColumWhitePoint(img, row, colum):
whitePoint = 0
for i in range(row):
if img[i, colum] > 0:
whitePoint += 1
return whitePoint
# 根据edge，垂直扫描白点数，没有则记录下最后一次没有白点的位置，记录下白点之后下一次没有白点的位置。
ToBlack = []
ToWhite = []
whitePoint = 0
nextHaveWhitePoint = False
nextHaveBlackPoint = False
for i in range(colum - 1):
whiteNumber = scanColumWhitePoint(numyPic, row, i)
# 这里需要修改，把处理过后的图片生成新的图片，处理黑行
if whiteNumber == 0:
if scanColumWhitePoint(numyPic, row, i + 1) > 0:
ToWhite.append(i+1)
if whiteNumber > 0:
if scanColumWhitePoint(numyPic, row, i + 1) == 0:
ToBlack.append(i + 1)
print("ToBlack: ", ToBlack)
# print("ToBlack: ", ToBlack[])
print("ToWhite: ", ToWhite)
for x in range(7):
cv2.imshow(str(x), numyPic[:, ToWhite[x]:ToBlack[x+1]])
# cv2.imshow('6', numyPic[:,45:53])
# cv2.imshow('aa', numyPic)
# cv2.imshow('a', binary)
# cv2.imshow('aaa', edge)
# cv2.imshow('aafa', imgg)
# element1 = cv2.getStructuringElement(cv2.MORPH_RECT, (9, 1))
# element2 = cv2.getStructuringElement(cv2.MORPH_RECT, (10, 3))
# # 膨胀一次，让轮廓突出
# dilation = cv2.dilate(binhull = cv2.convexHull(cnt)ary, element2, iterations=1)
# # 腐蚀一次，去掉细节
# erosion = cv2.erode(dilation, element1, iterations=1)
# # 再次膨胀，让轮廓明显一些
# dilation2 = cv2.dilate(erosion, element2, iterations=3)
# image, contours, hierarchy = cv2.findContours(dilation2, cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)
# imgg = cv2.drawContours(img, contours, -1, (0,255,255), 1)
# cv2.imshow('a', imgg)
# cv2.imshow('aa', dilation2)
cv2.waitKey(0)