def play_video(self):
"""Create a window and play the video stored in video_array.
generate_video will be run automatically if it has not been run.
Parameters
-----------
waitKey : int, optional
Time (ms) delay between showing each frame.
"""
temp_filename = self.temp_filename
window_name = temp_filename[(temp_filename.rfind('/') + 1):]
cv2.namedWindow(window_name, cv2.WINDOW_NORMAL)
cv2.resizeWindow(window_name, 640, 480)
print(
"'K' - Play/Pause | 'J' - Rewind 10 frames | 'L' - Fast Forward 10 frames | 'Q' - Quit"
)
frame_num = 0
pause = False
while True:
#TODO: Change progressbar to show timestamp (relative to video)
# instead of percentage
progress_bar.printProgressBar(frame_num,
self.num_frames - 1,
prefix='Playing Video: ')
key = cv2.waitKey(self.frame_delay)
if key == ord("q"):
break
elif key == ord("k"):
pause = not pause
elif key == ord("l"):
if frame_num < self.num_frames - 11:
frame_num += 10
else:
frame_num = self.num_frames - 1
elif key == ord("j"):
if frame_num > 10:
frame_num -= 10
else:
frame_num = 0
if not pause:
frame_num += 1
else:
pass
img = self.generate_frame(frame_num)
if frame_num == self.num_frames - 1:
frame_num = 0 #Start video over at end
cv2.imshow(window_name, img)
print(
'\n'
) # Print blank line to remove progressbar if video was quit before ending
cv2.destroyAllWindows()
def save_video(self, playback_speed=15, realtime_framerate=4):
"""Save the video as a .avi file.
Parameters
----------
playback_speed : int, optional
Multiple of realtime speed to playback video at.
realtime_framerate : int, optional
Number of frames taken by the thermal camera in real time.
"""
framerate = playback_speed * realtime_framerate
height = self.temp_data[0].shape[0] * self.scale_factor
width = self.temp_data[0].shape[1] * self.scale_factor
size = (width, height)
filename = asksaveasfilename()
video_writer = cv2.VideoWriter(
filename, cv2.VideoWriter_fourcc('F', 'M', 'P', '4'), framerate,
size)
for i, frame in enumerate(self.temp_data):
# Display completion percentage
progress_bar.printProgressBar(i,
self.num_frames,
prefix='Saving Video...')
img = self.generate_frame(i)
video_writer.write(img)
video_writer.release()
def save_hotspot_video(self,
playback_speed=15,
realtime_framerate=4,
save_img=False):
framerate = playback_speed * realtime_framerate
height = self.temp_data[0].shape[0] * self.scale_factor
width = self.temp_data[0].shape[1] * self.scale_factor
size = (width, height)
filename = asksaveasfilename()
video_writer = cv2.VideoWriter(
filename, cv2.VideoWriter_fourcc('F', 'M', 'P', '4'), framerate,
size)
hotspot_img = np.zeros((height, width), dtype=np.float32)
for i, frame in enumerate(self.temp_data, 0):
# Display completion percentage
progress_bar.printProgressBar(
i,
self.num_frames,
prefix='Saving Hotspot Video...',
)
current_max = np.amax(frame)
current_max_y = np.where(frame == current_max)[0][0]
current_max_x = np.where(frame == current_max)[1][0]
hotspot_img[current_max_y, current_max_x] = current_max
hotspot_img_frame = hotspot_img.copy()
hotspot_img_frame = cv2.normalize(
src=np.float32(hotspot_img_frame),
dst=np.float32(hotspot_img_frame),
alpha=0,
beta=255,
norm_type=cv2.NORM_MINMAX,
dtype=cv2.CV_8UC1)
hotspot_img_frame = cv2.applyColorMap(hotspot_img_frame,
cv2.COLORMAP_INFERNO)
video_writer.write(hotspot_img_frame)
video_writer.release()
if save_img:
cv2.imwrite('hotspot_img.png', hotspot_img_frame)
def generate_video(self):
"""Load temperature and meltpool data, match them, and create the video array"""
self.video_array = []
if self.remove_bottom_reflection:
lower_bounds = np_vid_viewer.reflection_remover.find_lower_bounds(
self.temp_data)
# Loop through each frame of data
i = 0
for frame in self.temp_data:
frame = frame.copy() # Make copy since file is read-only
progress_bar.printProgressBar(i,
self.num_frames,
prefix='Generating Video...')
if self.remove_top_reflection:
np_vid_viewer.reflection_remover.remove_top(
frame, zero_level_threshold=180, max_temp_threshold=700)
if self.remove_bottom_reflection:
np_vid_viewer.reflection_remover.remove_bottom(
frame, lower_bounds)
# Normalize the image to 8 bit color
img = frame.copy()
img = cv2.normalize(img, img, 0, 255, cv2.NORM_MINMAX, cv2.CV_8UC1)
# Apply colormap to image
img = cv2.applyColorMap(img, cv2.COLORMAP_INFERNO)
# Add meltpool data onto the image
if self.mp_data_on_vid:
self.add_mp_data_to_img(img, i)
# Add image to video array
self.video_array.append(img)
i = i + 1
def generate_threshold_image(self, threshold=800):
height = self.temp_data[0].shape[0] * self.scale_factor
width = self.temp_data[0].shape[1] * self.scale_factor
threshold_img = np.zeros((height, width), dtype=np.float32)
for i, frame in enumerate(self.temp_data, 0):
# Show progress
progress_bar.printProgressBar(
i, self.num_frames, prefix='Generating threshold image...')
# Check each pixel, if pixel is over threshold, increment that pixel in theshold_img
for y, row in enumerate(frame):
for x, pixel in enumerate(row):
if pixel > threshold:
threshold_img[y, x] += 1
threshold_img = cv2.normalize(src=threshold_img,
dst=threshold_img,
alpha=0,
beta=255,
norm_type=cv2.NORM_MINMAX,
dtype=cv2.CV_8UC1)
threshold_img = cv2.applyColorMap(threshold_img, cv2.COLORMAP_INFERNO)
cv2.imwrite("threshold_img.png", threshold_img)