def weightVideo(vidName):
vidFile = cv2.VideoCapture(vidName)
nFrames = int(vidFile.get(cv.CV_CAP_PROP_FRAME_COUNT))
fps = vidFile.get(cv.CV_CAP_PROP_FPS)
waitPerFrameInMillisec = int(1 / fps * 1000 / 1)
print 'Num. Frames = ', nFrames
print 'Frame Rate = ', fps, ' frames per sec'
s, background = vidFile.read()
shape = background.shape
background = cv2.cvtColor(background, cv2.COLOR_RGB2GRAY)
background = background / 255.0
mu = 0.01
for f in xrange(nFrames - 1):
s, frameImg = vidFile.read()
gray = cv2.cvtColor(frameImg, cv2.COLOR_RGB2GRAY)
gray = gray / 255.0
background = (1 - mu) * background + mu * gray
# for i in range(shape[0]):
# print i
# for j in range(shape[1]):
# z = background[i,j]
# background[i,j] = sigmoid(z*1.0)
# print background
# background=cv2.cvtColor(background,cv2.CV_8U)
cv2.imshow("Original Frame", utils.rotateImg(frameImg))
cv2.imshow("My Video Window", utils.rotateImg(background))
cv2.waitKey(waitPerFrameInMillisec)
def weightVideo(vidName): vidFile = cv2.VideoCapture(vidName) nFrames = int(vidFile.get(cv.CV_CAP_PROP_FRAME_COUNT) ) fps = vidFile.get(cv.CV_CAP_PROP_FPS ) waitPerFrameInMillisec = int( 1/fps * 1000/1 ) print 'Num. Frames = ', nFrames print 'Frame Rate = ', fps, ' frames per sec' s,background=vidFile.read() shape = background.shape background=cv2.cvtColor(background,cv2.COLOR_RGB2GRAY) background = background/255.0 mu = 0.01 for f in xrange( nFrames -1 ): s,frameImg = vidFile.read() gray = cv2.cvtColor(frameImg,cv2.COLOR_RGB2GRAY) gray = gray / 255.0 background = (1-mu) * background + mu * gray # for i in range(shape[0]): # print i # for j in range(shape[1]): # z = background[i,j] # background[i,j] = sigmoid(z*1.0) # print background # background=cv2.cvtColor(background,cv2.CV_8U) cv2.imshow( "Original Frame", utils.rotateImg(frameImg)) cv2.imshow( "My Video Window", utils.rotateImg(background)) cv2.waitKey( waitPerFrameInMillisec )
def flipPasteImage(self, paste_img, labels):
flip_val = np.random.randint(0, 100)
if flip_val < 50:
logging.info("Rotating by -90.")
rotate_deg = -90
else:
logging.info("Rotating by 90.")
rotate_deg = 90
rotated_paste_img, = utils.rotateImg(paste_img, rotate_deg)
new_labels = []
for l in labels:
new_right = utils.rotate([0.0, 0.0], l['x'], rotate_deg)
new_top = utils.rotate([0.0, 0.0], l['y'], rotate_deg)
new_left = utils.rotate([0.0, 0.0], l['x'] + l['width'],
rotate_deg)
new_bottom = utils.rotate([0.0, 0.0], l['y'] + l['height'],
rotate_deg)
new_l = l.copy()
new_l['x'] = new_right
new_l['y'] = new_top
new_l['width'] = new_left - new_right
new_l['height'] = new_bottom - new_top
new_labels.append(new_l)
self.printLabelDims(new_labels)
return rotated_paste_img, new_labels
def rotateCanvasImage(self, canvas_img, rotate_deg=0):
if rotate_deg == 0:
rotate_deg = self.getForcedRandomRotationValue(
self.max_canvas_rotation)
logging.info(" rotate_deg: {}.".format(rotate_deg))
rotated_canvas_img, rotated_canvas_mask = utils.rotateImg(canvas_img,
rotate_deg,
mask_in=None)
# utils.showAndWait('rotated_paste_img', rotated_canvas_img)
if rotated_canvas_img.shape[0] != self.final_img_height or \
rotated_canvas_img.shape[1] != self.final_img_width:
left_idx = int(rotated_canvas_img.shape[1] / 2.0 -
self.final_img_width / 2.0)
right_idx = left_idx + self.final_img_width
top_idx = int(rotated_canvas_img.shape[0] / 2.0 -
self.final_img_height / 2.0)
bottom_idx = top_idx + self.final_img_height
rotated_canvas_img = rotated_canvas_img[top_idx:bottom_idx,
left_idx:right_idx]
return rotated_canvas_img
# background *= 1./255
print background.shape
background = cv2.cvtColor(background, cv2.COLOR_RGB2GRAY)
background = background / 255.0
print background.shape
print np.amax(background)
mu = 0.05
for f in xrange(nFrames):
s, frameImg = vidFile.read()
if f % 10 == 0:
gray = cv2.cvtColor(frameImg, cv2.COLOR_RGB2GRAY)
retinexed = np.zeros([gray.shape[0], gray.shape[1], 3])
for i in range(2):
retinexed[:, :, i] = gray
retinexed = cca.retinex(retinexed)
retinexed = cv2.cvtColor(retinexed, cv2.cv.CV_BGR2GRAY)
cv2.imshow("retinexed frame", retinexed)
retinexed = retinexed / 255.0
background = (1 - mu) * background + mu * retinexed
# print background
# background=cv2.cvtColor(background,cv2.CV_8U)
cv2.imshow("Original Frame", utils.rotateImg(frameImg))
cv2.imshow("My Video Window", utils.rotateImg(background))
cv2.waitKey(waitPerFrameInMillisec)
# # When playing is done, delete the window
# # NOTE: this step is not strictly necessary,
# # when the script terminates it will close all windows it owns anyways
# cv2.DestroyWindow( "My Video Window" )
def rotatePasteImage(self, img, labels, rotate_deg=0):
# drawn_img = self.drawLabels(img.copy(), labels)
# utils.showAndWait('drawn_img', drawn_img)
rotated, mask = utils.rotateImg(img, rotate_deg)
# rotated = misc.imrotate(img, rotate_deg)
# utils.showAndWait('rotated', rotated)
center_x = img.shape[1] / 2.0
center_y = img.shape[0] / 2.0
new_center_x = rotated.shape[1] / 2.0
new_center_y = rotated.shape[0] / 2.0
new_labels = []
for l in labels:
w = l['width'] # * 0.85
h = l['height'] # * 0.85
tl = [l['x'] - center_x, (l['y'] - center_y)]
tr = [l['x'] + w - center_x, (l['y'] - center_y)]
bl = [l['x'] - center_x, (l['y'] + h - center_y)]
br = [l['x'] + w - center_x, (l['y'] + h - center_y)]
top_left = utils.rotate([0.0, 0.0], tl, -rotate_deg)
top_right = utils.rotate([0.0, 0.0], tr, -rotate_deg)
bottom_left = utils.rotate([0.0, 0.0], bl, -rotate_deg)
bottom_right = utils.rotate([0.0, 0.0], br, -rotate_deg)
tl2 = [(top_left[0] + new_center_x), (top_left[1] + new_center_y)]
tr2 = [(top_right[0] + new_center_x),
(top_right[1] + new_center_y)]
bl2 = [(bottom_left[0] + new_center_x),
(bottom_left[1] + new_center_y)]
br2 = [(bottom_right[0] + new_center_x),
(bottom_right[1] + new_center_y)]
# rotated = cv2.line(rotated,
# (int(tl2[0]), int(tl2[1])),
# (int(tr2[0]), int(tr2[1])),
# color=(0, 0, 255), thickness=3)
# rotated = cv2.line(rotated,
# (int(tr2[0]), int(tr2[1])),
# (int(br2[0]), int(br2[1])),
# color=(0, 0, 255), thickness=3)
# rotated = cv2.line(rotated,
# (int(br2[0]), int(br2[1])),
# (int(bl2[0]), int(bl2[1])),
# color=(0, 0, 255), thickness=3)
# rotated = cv2.line(rotated,
# (int(bl2[0]), int(bl2[1])),
# (int(tl2[0]), int(tl2[1])),
# color=(0, 0, 255), thickness=3)
# utils.showAndWait('rotated', rotated)
min_x = min(tl2[0], tr2[0], bl2[0], br2[0])
max_x = max(tl2[0], tr2[0], bl2[0], br2[0])
min_y = min(tl2[1], tr2[1], bl2[1], br2[1])
max_y = max(tl2[1], tr2[1], bl2[1], br2[1])
new_l = l.copy()
new_l['x'] = min_x
new_l['y'] = min_y
new_l['width'] = (max_x - min_x)
new_l['height'] = (max_y - min_y)
new_labels.append(new_l)
utils.printLabelDims(new_labels)
return rotated, new_labels
def addPastedImages(self, canvas_img_file, canvas_img, paste_img_files,
paste_label_dir, save_img_dir, save_label_dir,
canvas_idx, tile_idx, out_labels):
"""
Adds paste images to the canvas file and saves it and the labels.
:param canvas_img_file: canvas image filename to split
:param canvas_img: canvas image to split
:param paste_img_files: paste image files.
:param width_multiple: multiple of width to final image size.
:param height_multiple: multiple of height to final image size.
:param canvas_idx: canvas index.
:param tile_idx: tile index.
"""
canvas_width = canvas_img.shape[1]
canvas_height = canvas_img.shape[0]
num_pastes = np.random.randint(self.min_pasted_per_canvas,
self.max_pasted_per_canvas)
labels = []
if canvas_height != self.final_img_height or canvas_width != self.final_img_width:
logging.error(
"The canvas height for a paste add does not match the final image dimensions. Skipping image."
)
return
canvas_label = np.zeros([canvas_height, canvas_width, 3],
dtype=np.uint8)
logging.info("num_pastes: {}".format(num_pastes))
for past_idx in range(num_pastes):
paste_img_file_idx = self.paste_image_idx
# paste_img_file = '/media/dcofer/Ubuntu_Data/drone_images/drones/111.png'
paste_img_file = paste_img_files[paste_img_file_idx]
paste_filename = os.path.basename(paste_img_file)
paste_basename = os.path.splitext(paste_filename)[0]
logging.info(" Pasting in {}".format(paste_img_file))
logging.info(" Paste Image Idx {}".format(paste_img_file_idx))
paste_img = self.loadPasteImage(paste_img_file)
# utils.showAndWait('paste_img', paste_img)
paste_label_file = paste_label_dir + '/' + paste_basename + '_label.png'
paste_label_img = cv2.imread(paste_label_file,
cv2.IMREAD_UNCHANGED)
paste_label_img = cv2.cvtColor(paste_label_img, cv2.COLOR_BGR2GRAY)
# utils.showAndWait('paste_label_img', paste_label_img)
paste_width = paste_img.shape[1]
paste_height = paste_img.shape[0]
logging.info(" paste_width: {}".format(paste_width))
logging.info(" paste_height: {}".format(paste_height))
if paste_label_img.shape[
0] != paste_height or paste_label_img.shape[
1] != paste_width:
raise RuntimeError(
"Paste label dims do not match paste image.")
new_paste_width, new_paste_height = self.calcNewPasteDims(
paste_width, paste_height)
logging.info(" new_paste_width: {}".format(new_paste_width))
logging.info(" new_paste_height: {}".format(new_paste_height))
if paste_width != new_paste_width or paste_height != new_paste_height:
sized_paste_img = cv2.resize(paste_img,
dsize=(new_paste_width,
new_paste_height),
interpolation=cv2.INTER_AREA)
sized_paste_mask = cv2.resize(paste_label_img,
dsize=(new_paste_width,
new_paste_height),
interpolation=cv2.INTER_AREA)
else:
sized_paste_img = paste_img
sized_paste_mask = paste_label_img
# utils.showAndWait('sized_paste_img', sized_paste_img)
# utils.showAndWait('sized_paste_mask', sized_paste_mask)
flip_val = np.random.randint(0, 100)
if flip_val < 50:
logging.info(
" flip_val: {}. Flipping image.".format(flip_val))
flipped_paste_img = np.fliplr(sized_paste_img)
flipped_paste_mask = np.fliplr(sized_paste_mask)
else:
logging.info(
" flip_val: {}. Leaving image unflipped".format(
flip_val))
flipped_paste_img = sized_paste_img
flipped_paste_mask = sized_paste_mask
rotate_deg = int(
np.random.uniform(-self.max_paste_rotation,
self.max_paste_rotation))
logging.info(" rotate_deg: {}.".format(rotate_deg))
rotated_paste_img, rotated_paste_mask = utils.rotateImg(
flipped_paste_img, rotate_deg, mask_in=flipped_paste_mask)
# utils.showAndWait('rotated_paste_img', rotated_paste_img)
# utils.showAndWait('rotated_paste_mask', rotated_paste_mask)
paste_width = rotated_paste_img.shape[1]
paste_height = rotated_paste_img.shape[0]
paste_x, paste_y = self.getNonoverlappingPastePos(
paste_width, paste_height, canvas_width, canvas_height, labels)
if paste_x < 0 or paste_y < 0:
break
# paste_x = 1081
# paste_y = 266
logging.info(" paste_x: {}".format(paste_x))
logging.info(" paste_y: {}".format(paste_y))
# Get canvas ROI
canvas_roi = canvas_img[paste_y:(paste_y + paste_height),
paste_x:(paste_x + paste_width)]
#canvas_roi = np.zeros([paste_height, paste_width, 3], dtype=np.uint8)
# showAndWait('canvas_roi', canvas_roi)
# Regnerate a new mask because the one that was put through all the processing is not
# intact anymore. This was causing weird artifacting.
ret, mask = cv2.threshold(rotated_paste_mask, 5, 255,
cv2.THRESH_BINARY)
# mask = new_mask[:, :, 0]
mask_inv = cv2.bitwise_not(mask)
# utils.showAndWait('mask', mask)
# Black out the are of the mask.
background_roi = cv2.bitwise_and(canvas_roi,
canvas_roi,
mask=mask_inv)
# showAndWait('background_roi', background_roi)
# cv2.imwrite('/media/dcofer/Ubuntu_Data/drone_images/canvas_ros.png', background_roi)
# Now randomly change brightness and contract of foreground drone
bright_rand = np.random.randint(0, 100)
if bright_rand < self.blur_thresh:
logging.info(
" bright_rand: {}. Adjusting brightness/contrast.".
format(bright_rand))
bright_val = np.random.randint(-self.bright_max,
self.bright_max)
contrast_val = np.random.normal(1.0, self.contrast_max)
if contrast_val < 0.5:
contrast_val = 0.7
if contrast_val > 1.5:
contrast_val = 1.3
logging.info(" bright_val: {}".format(bright_val))
logging.info(" contrast_val: {}".format(contrast_val))
bright_foreground_img = cv2.convertScaleAbs(rotated_paste_img,
alpha=contrast_val,
beta=bright_val)
# utils.showAndWait('rotated_paste_img', rotated_paste_img)
# utils.showAndWait('bright_foreground_img', bright_foreground_img)
else:
logging.info(
" bright_rand: {}. Leaving image brightness/contrast alone"
.format(bright_rand))
bright_foreground_img = rotated_paste_img
# Now take only region of paste image that is not black
foreground_roi = cv2.bitwise_and(bright_foreground_img,
bright_foreground_img,
mask=mask)
# showAndWait('foreground_roi', foreground_roi)
# cv2.imwrite('/media/dcofer/Ubuntu_Data/drone_images/foreground_roi.png', foreground_roi)
# Put them together
merged_roi = cv2.add(background_roi, foreground_roi)
# showAndWait('merged_roi', merged_roi)
# cv2.imwrite('/home/mfp/drone-net/test/merged_roi.png', merged_roi)
# Now randomly add blur
blur_rand = np.random.randint(0, 100)
if blur_rand < self.blur_thresh:
logging.info(
" blur_rand: {}. bluring image.".format(blur_rand))
blur_val = np.random.randint(1, self.blur_max)
logging.info(" blur_val: {}".format(blur_val))
if blur_val > 0:
# blured_roi = cv2.GaussianBlur(merged_roi, (blur_val, blur_val), 0)
blured_roi = cv2.blur(merged_roi, (blur_val, blur_val))
else:
blured_roi = merged_roi
# cv2.imwrite('/home/mfp/drone-net/test/blured_roi.png', blured_roi)
else:
logging.info(
" blur_rand: {}. Leaving image un-blurred".format(
blur_rand))
blured_roi = merged_roi
# Now put them back into the canvas
canvas_img[paste_y:(paste_y + paste_height),
paste_x:(paste_x + paste_width)] = blured_roi
# utils.showAndWait('canvas_img', canvas_img)
# Now put the label into the canvas
ret, label_mask = cv2.threshold(rotated_paste_mask, 5, 255,
cv2.THRESH_BINARY)
where_label = np.array(np.where(label_mask))
canvas_label_roi = canvas_label[paste_y:(paste_y +
label_mask.shape[0]),
paste_x:(paste_x +
label_mask.shape[1]), 2]
canvas_label_roi[where_label[0], where_label[1]] = 255
canvas_label[paste_y:(paste_y + label_mask.shape[0]),
paste_x:(paste_x + label_mask.shape[1]),
2] = canvas_label_roi
# utils.showAndWait('canvas_label', canvas_label)
self.canvas_paste_links.append([
canvas_idx, tile_idx, canvas_img_file, paste_img_file, paste_x,
paste_y, paste_width, paste_height
])
json_label = {
"class": "rect",
"height": paste_height,
"width": paste_width,
"x": paste_x,
"y": paste_y
}
labels.append(json_label)
self.incrementNextPastedImageIndex(paste_img_files)
# canvas_img = utils.drawLabels(canvas_img, labels)
# utils.showAndWait('canvas_img', canvas_img)
save_img_file = save_img_dir + '/{}_{}_{}.jpg'.format(
self.file_prefix, canvas_idx, tile_idx)
cv2.imwrite(save_img_file, canvas_img)
logging.info("saving image: {}".format(save_img_file))
#misc.imsave(save_file, canvas_img)
save_label_file = save_label_dir + '/{}_{}_{}.txt'.format(
self.file_prefix, canvas_idx, tile_idx)
utils.saveYoloLabelFile(0, labels, save_label_file, canvas_width,
canvas_height)
logging.info("saving lable: {}".format(save_label_file))
#
# save_label_file = save_label_dir + '/()_{}_{}_label.png'.format(self.file_prefix, canvas_idx, tile_idx)
# cv2.imwrite(save_label_file, canvas_label)
# logging.info("saving lable image: {}".format(save_label_file))
json_label = {
"class": "image",
"filename": save_img_file,
"annotations": labels
}
out_labels.append(json_label)