-
Notifications
You must be signed in to change notification settings - Fork 0
/
mosaic.py
173 lines (120 loc) · 5.51 KB
/
mosaic.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
import cv2
import numpy as np
import matplotlib.pyplot as plt
##############################################################
### FUNCTIONS
##############################################################
### 1. Extract ORB keypoints and descriptors from a gray image
def extract_features(gray):
## TODO: Detect ORB features and compute descriptors.
## TODO: (Overwrite the following 2 lines with your answer.)
orb = cv2.ORB_create()
kp, des = orb.detectAndCompute(gray, None)
return (kp, des)
### 2. Find corresponding features between the images
def find_matches(keypoints1, descriptors1, keypoints2, descriptors2):
## TODO: Look up corresponding keypoints.
## TODO: (Overwrite the following 2 lines with your answer.)
bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)
# Match descriptors
matches = bf.match(descriptors1, descriptors2)
# Sort them in the order of their distance.
matches = sorted(matches, key=lambda x: x.distance)
# convert first 10 matches from KeyPoint objects to NumPy arrays
points1 = np.float32([kp1[m.queryIdx].pt for m in matches[0:50]])
points2 = np.float32([kp2[m.trainIdx].pt for m in matches[0:50]])
return (points1, points2)
### 3. Find homography between the points
def find_homography(points1, points2):
# convert the keypoints from KeyPoint objects to NumPy arrays
src_pts = points2.reshape(-1,1,2)
dst_pts = points1.reshape(-1,1,2)
# find homography
homography, mask = cv2.findHomography(src_pts, dst_pts)
return (homography)
### 4.1 Calculate the size and offset of the stitched panorama
def calculate_size(image1, image2, H):
# compute de coordinates of image2 corners in image1
h2 = image2.shape[0]
w2 = image2.shape[1]
corners2 = np.float32([[[0, 0], [0, h2 - 1], [w2 - 1, h2 - 1], [w2 - 1, 0]]])
transformedCorners2 = cv2.perspectiveTransform(corners2, H)
pointsImage1=([0,0],[0,image1.shape[0]],[image1.shape[1],0],[image1.shape[1],image1.shape[0]])
total=np.concatenate((transformedCorners2[0], pointsImage1), axis=0)
print(total)
plt.plot(total[0:4,0],total[0:4,1], 'ro')
plt.plot(total[4:8, 0], total[4:8, 1], 'go')
plt.xlabel('green first image, red second transformed')
plt.axis([-1500, w2+3000, -1500, h2+3000])
maxX=int(round(max(total[:,0])))
minX=int(round(min(total[:,0])))
maxY=int(round(max(total[:,1])))
minY=int(round(min(total[:,1])))
offset = (np.abs(minX), np.abs(minY))
#cv2.waitKey(0)
## Update the homography to shift by the offset
size= (int(round(maxX-minX)),int(round(maxY-minY)))
H[0:2, 2] += offset
return (size, offset)
## 4.2 Combine images into a panorama
def merge_images(image1, image2, homography, size, offset):
img2 = cv2.warpPerspective(image2, homography, size)
## TODO: Combine the two images into one.
## TODO: (Overwrite the following 5 lines with your answer.)
(h1, w1) = image1.shape[:2]
(hwarp, wwarp) = img2.shape[:2]
panorama = np.zeros((size[1], size[0], 3), np.uint8)
panorama[:hwarp, :wwarp] = img2
panorama[offset[1]:h1+offset[1], offset[0]:w1+offset[0]] = image1
img2=cv2.resize(img2, (0, 0), fx=0.3, fy=0.3)
cv2.imshow('hist', img2)
return panorama
### --- No need to change anything below this point ---
### Connects corresponding features in the two images using yellow lines
def draw_matches(image1, image2, points1, points2):
# Put images side-by-side into 'image'
(h1, w1) = image1.shape[:2]
(h2, w2) = image2.shape[:2]
image = np.zeros((max(h1, h2), w1 + w2, 3), np.uint8)
image[:h1, :w1] = image1
image[:h2, w1:w1 + w2] = image2
# Draw yellow lines connecting corresponding features.
for (x1, y1), (x2, y2) in zip(np.int32(points1), np.int32(points2)):
cv2.line(image, (x1, y1), (x2 + w1, y2), (0, 255, 255))
return image
##############################################################
### MAIN PROGRAM
##############################################################
### Load images
img1 = cv2.imread('C://Users/gecete/Documents/ESTER/Image1.jpg')
img2 = cv2.imread('C://Users/gecete/Documents/ESTER/Image2.jpg')
# Convert images to grayscale (for ORB detector).
gray1 = cv2.cvtColor(img1, cv2.COLOR_BGR2GRAY)
gray2 = cv2.cvtColor(img2, cv2.COLOR_BGR2GRAY)
### 1. Detect features and compute descriptors.
(kp1, desc1) = extract_features(img1)
(kp2, desc2) = extract_features(img2)
print ('{0} features detected in image1').format(len(kp1))
print ('{0} features detected in image2').format(len(kp2))
orb1 = cv2.drawKeypoints(gray1, kp1, None, flags=cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS)
orb2 = cv2.drawKeypoints(gray2, kp2, None, flags=cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS)
cv2.imwrite('Image1_orf.JPG', orb1)
cv2.imwrite('Image2_orb.JPG', orb2)
cv2.imshow('Features 1', orb1)
cv2.imshow('Features 2', orb2)
### 2. Find corresponding features
(points1, points2) = find_matches(kp1, desc1, kp2, desc2)
print ('{0} features matched').format(len(points1))
match = draw_matches(img1, img2, points1, points2)
cv2.imwrite('matching.JPG', match)
cv2.imshow('Matching', match)
### 3. Find homgraphy
H = find_homography(points1, points2)
### 4. Combine images into a panorama
(size, offset) = calculate_size(img1, img2, H)
print ('output size: {0} offset: {1}').format(size, offset)
panorama = merge_images(img1, img2, H, size, offset)
cv2.imwrite("panorama.jpg", panorama)
cv2.imshow('Panorama', panorama)
plt.show()
cv2.waitKey(0)