-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathBirdsEyePerspectiveTransformationUtility.py
170 lines (137 loc) · 6.26 KB
/
BirdsEyePerspectiveTransformationUtility.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
# Quick and dirty utility to get coordinates for transforming view into
# a bird's eye view. Useful in OCRs were the camera is in a fixed positioning
# viewing a straight plane.
import cv2
import numpy as np
def onTrackbarChange(trackbarValue):
pass
def order_points(pts):
# initialize a list of coordinates that will be ordered
# such that the first entry in the list is the top-left,
# the second entry is the top-right, the third is the
# bottom-right, and the fourth is the bottom-left
rect = np.zeros((4, 2), dtype = "float32")
# the top-left point will have the smallest sum, whereas
# the bottom-right point will have the largest sum
s = pts.sum(axis = 1)
rect[0] = pts[np.argmin(s)]
rect[2] = pts[np.argmax(s)]
# now, compute the difference between the points, the
# top-right point will have the smallest difference,
# whereas the bottom-left will have the largest difference
diff = np.diff(pts, axis = 1)
rect[1] = pts[np.argmin(diff)]
rect[3] = pts[np.argmax(diff)]
# return the ordered coordinates
return rect
def expandPerspective(rect , width, height):
'''Expand the perspective out to the image limits
by finding intersection using point-slope form'''
# Constants
x = 0
y = 1
# Convert coordinate system
rect[:,1] *= -1
(tl, tr, br, bl) = rect
# Find the slope of each of the 4 lines
slopeTop = (tr[y]-tl[y]) / (tr[x]-tl[x])
slopeBottom = (br[y]-bl[y]) / (br[x]-bl[x])
slopeLeft = (tl[y]-bl[y]) / (tl[x]-bl[x])
slopeRight = (tr[y]-br[y]) / (tr[x]-br[x])
# Assign new points based on image size
pointRight = width,0
pointTop = 0,0
pointBottom = width, height * -1.0
pointLeft = 0, height* -1.0
# Find where the new expanded lines intersect using point slope form
def intersectoin (m1,m2,x1,x2,y1,y2,orig):
x = ((m2*x2-m1*x1)-(y2-y1))/(m2-m1)
#y = ((-1.0*m1*y2 + m1*m2*x2 + y1*m2 )-(m1*m2*x1))/(m2-m1)
y = m1*(x - x1) + y1
try:
x = round(x)
y = round(y)
except:
return orig
return x, y
new_tr = intersectoin (slopeTop,slopeRight,pointTop[x],pointRight[x],pointTop[y],pointRight[y],tr)
new_tl = intersectoin (slopeTop,slopeLeft,pointTop[x],pointLeft[x],pointTop[y],pointLeft[y],tl)
new_br = intersectoin (slopeBottom,slopeRight,pointBottom[x],pointRight[x],pointBottom[y],pointRight[y],br)
new_bl = intersectoin (slopeBottom,slopeLeft,pointBottom[x],pointLeft[x],pointBottom[y],pointLeft[y],bl)
# Convert coordinate system back
new_rect = rect = np.array([new_tl, new_tr, new_br, new_bl], dtype = "float32")
new_rect[:,1] *= -1
return new_rect
# Derived from https://www.pyimagesearch.com/2014/08/25/4-point-opencv-getperspective-transform-example
def four_point_transform(image, pts):
# Unpack points
rect = pts
(tl, tr, br, bl) = rect
# compute the width of the new image, which will be the
# maximum distance between bottom-right and bottom-left
# x-coordinates or the top-right and top-left x-coordinates
widthA = np.sqrt(((br[0] - bl[0]) ** 2) + ((br[1] - bl[1]) ** 2))
widthB = np.sqrt(((tr[0] - tl[0]) ** 2) + ((tr[1] - tl[1]) ** 2))
maxWidth = max(int(widthA), int(widthB))
# compute the height of the new image, which will be the
# maximum distance between the top-right and bottom-right
# y-coordinates or the top-left and bottom-left y-coordinates
heightA = np.sqrt(((tr[0] - br[0]) ** 2) + ((tr[1] - br[1]) ** 2))
heightB = np.sqrt(((tl[0] - bl[0]) ** 2) + ((tl[1] - bl[1]) ** 2))
maxHeight = max(int(heightA), int(heightB))
# now that we have the dimensions of the new image, construct
# the set of destination points to obtain a "birds eye view",
# (i.e. top-down view) of the image, again specifying points
# in the top-left, top-right, bottom-right, and bottom-left
# order
dst = np.array([
[0, 0], #tl
[maxWidth - 1, 0], #tr
[maxWidth - 1, maxHeight - 1], #br
[0, maxHeight - 1]], #bl
dtype = "float32")
# Move image to positive coordinates
min_x = round(abs(np.min(rect[:,0])))
min_y = round(abs(np.min(rect[:,1])))
T = np.matrix( [[ 1 , 0 , min_x], # Get min x
[ 0 , 1 , min_y ], # Get min y
[ 0 , 0 , 1 ]],
dtype = "float32")
# compute the perspective transform matrix and then apply it
M = cv2.getPerspectiveTransform(rect, dst)
warped = cv2.warpPerspective(image, T * M , (maxWidth + min_x , maxHeight + min_y), borderMode=cv2.BORDER_TRANSPARENT)
# return the warped image
return warped
# Open Image
img = cv2.imread('img\\example1.jpeg')
# Open windows for control, original image, and result
cv2.namedWindow('Control', cv2.WINDOW_AUTOSIZE)
cv2.namedWindow('Main', cv2.WINDOW_AUTOSIZE)
cv2.namedWindow('Birds Eye', cv2.WINDOW_AUTOSIZE)
# Track bars for coordinates
cv2.createTrackbar( 'X L Bot', 'Control', 0, img.shape[1], onTrackbarChange )
cv2.createTrackbar( 'Y L Bot', 'Control', img.shape[0], img.shape[0], onTrackbarChange )
cv2.createTrackbar( 'X L Top', 'Control', 0, img.shape[1], onTrackbarChange )
cv2.createTrackbar( 'Y L Top', 'Control', 0, img.shape[0], onTrackbarChange )
cv2.createTrackbar( 'X R Top', 'Control', img.shape[1], img.shape[1], onTrackbarChange )
cv2.createTrackbar( 'Y R Top', 'Control', 0, img.shape[0], onTrackbarChange )
cv2.createTrackbar( 'X R Bot', 'Control', img.shape[1], img.shape[1], onTrackbarChange )
cv2.createTrackbar( 'Y R Bot', 'Control', img.shape[0], img.shape[0], onTrackbarChange )
# Loop
while(1):
# Get Track Bar positions
pts = np.array(eval('[(' + str(cv2.getTrackbarPos('X L Bot','Control')) + ',' + str(cv2.getTrackbarPos('Y L Bot','Control')) + '),' +
'(' + str(cv2.getTrackbarPos('X L Top','Control')) + ',' + str(cv2.getTrackbarPos('Y L Top','Control'))+ '),' +
'(' + str(cv2.getTrackbarPos('X R Top','Control')) + ',' + str(cv2.getTrackbarPos('Y R Top','Control'))+ '),' +
'(' + str(cv2.getTrackbarPos('X R Bot','Control')) + ',' + str(cv2.getTrackbarPos('Y R Bot','Control'))+ ')]'
), dtype = "int32")
# Draw the perspective
imgConnectedPoints = cv2.polylines(img.copy(), [pts], isClosed = True, color = (0,255,0), thickness = 3)
cv2.imshow('Main',imgConnectedPoints)
# Draw the transformed bird's eye view
warped = four_point_transform(img, expandPerspective(order_points(pts), img.shape[1], img.shape[0]))
cv2.imshow('Birds Eye',warped)
# Exit
if cv2.waitKey(1)==27:
exit(0)
cv.detroyAllWindows()