-
Notifications
You must be signed in to change notification settings - Fork 7
/
camera.py
305 lines (261 loc) · 10.4 KB
/
camera.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
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
# **************************************************************************** #
# #
# ::: :::::::: #
# camera.py :+: :+: :+: #
# +:+ +:+ +:+ #
# By: taston <thomas.aston@ed.ac.uk> +#+ +:+ +#+ #
# +#+#+#+#+#+ +#+ #
# Created: 2023/04/25 15:41:23 by taston #+# #+# #
# Updated: 2023/09/01 13:41:21 by taston ### ########.fr #
# #
# **************************************************************************** #
import numpy as np
import cv2
from datetime import datetime
from video import Video
class Camera:
"""
A class used to represent a Camera.
...
Attributes
----------
focal_length : float
float representing the focal length of the Camera in mm
internal_matrix : ndarray
array representing the Camera's intrinsic parameters
distortion_matrix : ndarray
array representing the Camera's lens distortion parameters
calibrator : Calibrator
Calibrator object used for camera calibration
calibrated : bool
bool for quick checking if camera has been calibrated
video : Video
Video object where the footage has been shot using this Camera
Methods
-------
calibrate(checkerboard=(9,6), video=Video()):
Performs calibration on the camera
"""
def __init__(self):
"""
Parameters
----------
...
"""
width = 1280
height = 720
self.focal_length = height * 1.28
# self.focal_length = 5000
self.internal_matrix = np.array([[self.focal_length, 0, width/2],
[0, self.focal_length, height/2],
[0, 0, 1]])
self.distortion_matrix = np.zeros((4, 1), dtype=np.float64)
self.calibrated = False
def calibrate(self, checkerboard=(9,6), video=Video(), show=True):
"""Creates a calibrator object and calibrates the Camera.
If arguments checkerboard and video aren't passed in, the
default checkerboard pattern and an empty video are used.
Parameters
----------
checkerboard : tuple, optional
Checkerboard pattern used in camera calibration (default is 9x6)
video : Video, optional
Video used to calibrate camera
"""
self.video = video
self.calibrator = Calibrator(checkerboard, self.video, show)
self.calibrated = True
self.internal_matrix, self.distortion_matrix = self.calibrator.matrix, self.calibrator.distortion
return self
class Checkerboard:
"""
A class used to represent a calibration Checkerboard
...
Attributes
----------
dimensions : tuple
tuple of checkerboard pattern dimensions
min_points : int
integer threshold of minimum detected points for
checkerboard to be considered found
objectp3d : ndarray
array of checkerboard points in three dimensions
threedpoints : list
list of checkerboard points in three dimensions
for each frame where a checkerboard is found
twodpoints : list
list of detected checkerboard points in two
dimensions for each frame
Methods
-------
get_corners(gray_frame)
Finds checkerboard corners in a given grayscale frame
"""
def __init__(self, dimensions = (9,6)):
"""
Parameters
----------
dimensions : tuple, optional
Checkerboard pattern used in camera calibration (default is 9x6)
"""
print('Checkerboard created')
self.dimensions = dimensions
self.min_points = 50
self.twodpoints = []
self.threedpoints = []
self.objectp3d = np.zeros((1, self.dimensions[0]
* self.dimensions[1],
3), np.float32)
self.objectp3d[0, :, :2] = np.mgrid[0:self.dimensions[0],
0:self.dimensions[1]].T.reshape(-1, 2)
def get_corners(self, gray_frame):
"""
Looks for checkerboard corners in a given grayscale video
frame.
Parameters
----------
gray_frame : ndarray
ndarray representing grayscale frame from video
Returns
-------
ret : bool
bool representing if corner search was successful
corners : ndarray
ndarray containing coordinates of corners
"""
ret, corners = cv2.findChessboardCorners(
gray_frame, self.dimensions,
cv2.CALIB_CB_ADAPTIVE_THRESH
+ cv2.CALIB_CB_FAST_CHECK +
cv2.CALIB_CB_NORMALIZE_IMAGE)
return ret, corners
class Calibrator:
"""
A class used to represent a camera Calibrator
...
Attributes
----------
checkerboard : Checkerboard
criteria : tuple
tuple of criteria for successful camera calibration
distortion : ndarray
ndarray of distortion parameters
frame : ndarray
ndarray representing video frame
gray_frame : ndarray
ndarray representing grayscale video frame
matrix : ndarray
ndarray representing camera intrinsic matrix
r_vecs : ndarray
ndarray of rotational vectors
t_vecs : ndarray
ndarray of translation vectors
Methods
-------
calibrate()
Perform camera calibration process
draw_corners(corners)
Draw checkerboard corners on video frame
save_outputs()
Save camera parameters to csv files
"""
def __init__(self, checkerboard, video=Video(), show=True):
"""
Parameters
----------
checkerboard : tuple
tuple representing Checkerboard pattern
video : Video, optional
Video used for camera calibration. If no video specified
an empty video will be attempted.
"""
timestamp = datetime.now().strftime("%H:%M:%S")
self.show = show
self.video = video
print('-'*120)
print('{:<100} {:>19}'.format(f'Creating Calibrator object for video {self.video.filename}:', timestamp))
print('-'*120)
print(self.video)
self.checkerboard = Checkerboard(checkerboard)
print(f'Checkerboard dimensions: {self.checkerboard.dimensions[0]} x {self.checkerboard.dimensions[1]}')
self.criteria = (cv2.TERM_CRITERIA_EPS +
cv2.TERM_CRITERIA_MAX_ITER, 30, 0.001)
self.calibrate()
# self.save_outputs()
timestamp = datetime.now().strftime("%H:%M:%S")
print('-'*120)
print('{:<100} {:>19}'.format(f'Calibrator object complete!', timestamp))
print('-'*120)
def calibrate(self):
"""
Performs the camera calibration procedure outlined here:
https://docs.opencv.org/4.x/dc/dbb/tutorial_py_calibration.html
"""
timestamp = datetime.now().strftime("%H:%M:%S")
self.video.create_writer()
print('Displaying video...')
while True:
ret, self.frame = self.video.cap.read()
frame_number = int(self.video.cap.get(cv2.CAP_PROP_POS_FRAMES))
self.gray_frame = cv2.cvtColor(self.frame, cv2.COLOR_BGR2GRAY)
ret, corners = self.checkerboard.get_corners(self.gray_frame)
if ret:
complete, image = self.draw_corners(corners)
if complete: break
if self.show == True:
cv2.imshow('Calibrating...', self.frame)
self.video.writer.write(self.frame)
k = cv2.waitKey(1)
if k == 27:
self.video.cap.release()
self.video.writer.release()
cv2.destroyAllWindows()
break
h, w = image.shape[:2]
# Perform camera calibration by given threedpoints and twodpoints
ret, self.matrix, self.distortion, self.r_vecs, self.t_vecs = cv2.calibrateCamera(self.checkerboard.threedpoints,
self.checkerboard.twodpoints,
self.gray_frame.shape[::-1], None, None)
print(f'Number of frames used for calibration: {frame_number}')
return self
def draw_corners(self, corners):
'''
Draws corners of checkerboard onto frame to verify calibration is working
Parameters
----------
corners : ndarray
ndarray of the corners found for a given frame
Returns
-------
complete : bool
bool representing whether search for corners is complete
frame : ndarray
new video frame with corners drawn
'''
complete = False
self.checkerboard.threedpoints.append(self.checkerboard.objectp3d)
# Refining pixel coordinates for given 2d points.
corners2 = cv2.cornerSubPix(
self.gray_frame, corners, self.checkerboard.dimensions, (-1, -1), self.criteria)
self.checkerboard.twodpoints.append(corners2)
# When we have minimum number of data points, stop:
if len(self.checkerboard.twodpoints) > self.checkerboard.min_points:
self.video.cap.release()
self.video.writer.release()
cv2.destroyAllWindows()
complete=True
# Draw and display the corners:
frame = cv2.drawChessboardCorners(self.frame,
self.checkerboard.dimensions,
corners2, True)
return complete, frame
def save_outputs(self):
"""
Saves matrices to csv
"""
timestamp = datetime.now().strftime("%H:%M:%S")
print('Saving outputs...')
from numpy import savetxt
savetxt('camera_matrix.csv', self.matrix, delimiter=',')
savetxt('camera_distortion.csv', self.distortion, delimiter=',')
return