def flippingOp(self):
# read image
im = cv2.imread(self.Image)
# flip the image horizontally
flipped = cv2.flip(im, 1)
cv2.imshow("Flipped Horizontally", flipped)
(b, g, r) = flipped[235, 259]
print("red=%d, green=%d, blue=%d" % (r,g,b))
cv2.waitKey(0)
# flip the image vertically
flipped = cv2.flip(im, 0)
cv2.imshow("Flipped Vertically", flipped)
cv2.waitKey(0)
# flip on both axis
flipped = cv2.flip(im, -1)
cv2.imshow("Flipped on both axis", flipped)
cv2.waitKey(0)
flipped = cv2.flip(im, 1)
cv2.imshow("1", flipped)
cv2.waitKey(0)
rotated = imutils.rotate(flipped, 45)
cv2.imshow("2", rotated)
cv2.waitKey(0)
flippedAgain = cv2.flip(rotated, 0)
cv2.imshow("3", flippedAgain)
cv2.waitKey(0)
(b, g, r) = flippedAgain[189, 441]
print("red=%d, green=%d, blue=%d" % (r,g,b))
return
def rotate_img(img, file_name, store_path, iter_count = 1): img_left_five = imutils.rotate(img, 3) img_left_five = cv2.resize(img_left_five, dim, interpolation=cv2.INTER_AREA) iter_count += 1 cv2.imwrite(join(store_path, file_name + '_' + str(iter_count)) + '.jpg', img_left_five) img_right_five = imutils.rotate(img, 358) img_right_five = cv2.resize(img_right_five, dim, interpolation=cv2.INTER_AREA) iter_count += 1 cv2.imwrite(join(store_path, file_name + '_' + str(iter_count) + '.jpg'), img_right_five) img_left_ten = imutils.rotate(img, 10) img_left_ten = cv2.resize(img_left_ten, dim, interpolation=cv2.INTER_AREA) iter_count += 1 cv2.imwrite(join(store_path, file_name + '_' + str(iter_count)) + '.jpg', img_left_ten) img_right_ten = imutils.rotate(img, 350) img_right_ten = cv2.resize(img_right_ten, dim, interpolation=cv2.INTER_AREA) iter_count += 1 cv2.imwrite(join(store_path, file_name + '_' + str(iter_count)) + '.jpg', img_right_ten)
def process_motion_frame(q, f, tick, ts, mfa=False, rotateAng=False, width=False, gBlur=(9, 9)):
'''
This function defines the image processing techniques that are applied
to a new thread when a frame is retreived from the camera.
'''
rects_sal = []
fgmask = None
f_copy = f.copy()
if rotateAng is not False and rotateAng != 0:
f = imutils.rotate(f, angle=rotateAng)
if width is not False:
f = imutils.resize(f, width=width)
# blur & bg sub
try:
fgmask = fgbg.apply(cv2.GaussianBlur(f, gBlur, 0), learningRate=config.computing.learning_rate)
except:
print("-"*60)
traceback.print_exc(file=sys.stdout)
print("-"*60)
raise
# get our frame outlines
f_rects, rects_mot = get_motions(f, fgmask, thickness=1)
rects_sal.extend(rects_mot)
num_motion = len(rects_mot)
if True:
# don't do anything else if there's no motion of any kind detected
# if num_motion > 0 or mfa is True:
num_bodies = 0
num_faces = 0
if config.computing.body_detection_en or config.computing.face_detection_en:
# generate a histogram equalized bw image if we're doing processing
# that needs it
f_bw = cv2.equalizeHist(cv2.cvtColor(f, cv2.COLOR_BGR2GRAY))
if config.computing.body_detection_en:
fBody, rectsBody = detectPerson(f, color=(255, 0, 0))
if len(rectsBody) > 0:
f_rects = cv2.add(f_rects, fBody)
num_bodies = len(rectsBody)
rects_sal.extend(rectsBody)
if config.computing.face_detection_en:
fFace, rectsFace = detectFace(f_bw, color=(0, 255, 0))
if len(rectsFace) > 0:
f_rects = cv2.add(f_rects, fFace)
num_faces = len(rectsFace)
rects_sal.extend(rectsFace)
f_rects = imutils.resize(f_rects, width=f_copy.shape[1])
q.put({"f": f_copy, "ts": ts, "rects_sal": rects_sal, "sz_scaled": getsize(
f), "num_motion": num_motion, "num_bodies": num_bodies, "num_faces": num_faces})
return f_copy, f_rects, fgmask, rects_sal, tick, ts
def rotationOp(self):
# read image
im = cv2.imread(self.Image)
# grab the dimension of the image
(h, w) = im.shape[:2]
(cX, cY) = (w/2, h/2)
# rotate image by 45 degrees
M = cv2.getRotationMatrix2D((cX, cY), 45, 1.0)
rotated = cv2.warpAffine(im, M, (w,h))
cv2.imshow("Rotated by 45 degrees", rotated)
cv2.waitKey(0)
# rotate our image by -90 degrees
M = cv2.getRotationMatrix2D((cX, cY), -90, 1.0)
rotated = cv2.warpAffine(im, M, (w, h))
cv2.imshow("Rotated by -90 Degrees", rotated)
cv2.waitKey(0)
# using imutils
rotated = imutils.rotate(im, 180)
cv2.imshow("Rotated by 180 degrees using imutils", rotated)
cv2.waitKey(0)
rotated = imutils.rotate(im, -110)
(b, g, r) = rotated[136, 312]
print("red=%d, green=%d, blue=%d" % (r,g,b))
M = cv2.getRotationMatrix2D((50, 50), -88, 1.0)
rotated = cv2.warpAffine(im, M, (w, h))
cv2.imshow("Rotated by -88 Degrees with offset (50,50)", rotated)
(b, g, r) = rotated[10, 10]
print("red=%d, green=%d, blue=%d" % (r,g,b))
cv2.waitKey(0)
return
import argparse
import cv2
import imutils as imu
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required=True, help="Path to image")
args = vars(ap.parse_args())
image = cv2.imread(args["image"])
print(image.shape)
cv2.imshow("Original", image)
(h, w) = image.shape[:2]
center = (w // 2, h // 2)
M = cv2.getRotationMatrix2D(center, 45, 1.0)
rotated = cv2.warpAffine(image, M, (w, h))
cv2.imshow("Rotated by 45 degrees", rotated)
M = cv2.getRotationMatrix2D(center, -90, 1.0)
rotated = cv2.warpAffine(image, M, (w, h))
cv2.imshow("Rotated by -90 degrees", rotated)
rotated = imu.rotate(image, 180)
cv2.imshow("Rotated by 180 degrees", rotated)
rotated = imu.rotate(image, 45)
cv2.imshow("Rotated by 45 degrees, imutil", rotated)
cv2.waitKey(0)
def getYlabel(img, xaxis):
label_images = []
ylabels = []
ypixels = []
img = img[:,:max(0,xaxis['start']-5)]
# Eliminate y title and tick marks
img = axis_processing.eliminateYTitle(img)
img = axis_processing.eliminateYTicks(img)
# Obtain labels
height, width = img.shape
blur = cv2.GaussianBlur(img, (7,7), 0)
thresh = cv2.threshold(blur, 0, 255, cv2.THRESH_BINARY_INV + cv2.THRESH_OTSU)[1]
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (2,1))
dilate = cv2.dilate(thresh, kernel, iterations=4)
cnts = cv2.findContours(dilate, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
for c in cnts:
x,y,w,h = cv2.boundingRect(c)
# cv2.rectangle(img, (x, y), (x + w, y + h), 0, 2)
ypixels.append(y+h//2)
margin = 5
label_images.append(img[max(0,y-margin):min(height,y+h+margin),
max(0,x-margin):min(width,x+w+margin)])
ylabels = [float('inf')]*len(ypixels)
for i,lab in enumerate(label_images):
lab = cv2.resize(lab, None, fx=10.5, fy=10.5, interpolation=cv2.INTER_CUBIC)
kernel = np.ones((1, 1), np.uint8)
lab = cv2.dilate(lab, kernel, iterations=4)
_, lab = cv2.threshold(lab, 200, 255, cv2.THRESH_BINARY)
lab = cv2.GaussianBlur(lab,(5,5),0)
text = pytesseract.image_to_string(lab, lang="eng", config="--psm 6 digits")
# print(f"Label{i}: {text}")
try:
ylabels[i] = float(text)
except:
pass
label_images[i] = lab
# cv2.imshow(f'label{i}', lab)
if (ylabels.count(float('inf')) > len(ylabels) - 3):
# Not enough points to perform linear interpolation
# We suspect that the numbers are rotated 90 deg CCW
for i,lab in enumerate(label_images):
lab = imutils.rotate(lab, -90)
lab = helper_functions.cropBlackBorder(lab)
text = pytesseract.image_to_string(lab, lang="eng", config="--psm 6 digits")
# print(f"Label{i}: {text}")
try:
ylabels[i] = float(text)
except:
pass
label_images[i] = lab
# cv2.imshow(f'label{i}', lab)
# cv2.imshow('thresh', thresh)
# cv2.imshow('dilate', dilate)
# cv2.imshow('image', img)
cv2.waitKey(0)
cv2.destroyAllWindows()
zipped_y = zip(ypixels, ylabels)
return list(zipped_y)
import numpy as np
import argparse
import imutils
import cv2
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required=True, help="Path to image")
args = vars(ap.parse_args())
image = cv2.imread(args["image"])
cv2.imshow("Original", image)
(h, w) = image.shape[:2]
center = (w // 2, h // 2)
M = cv2.getRotationMatrix2D(center, 45, 1.0)
rotated = cv2.warpAffine(image, M, (w, h))
cv2.imshow("Rotated by 45 degree", rotated)
M = cv2.getRotationMatrix2D(center, -90, 1.0)
rotated = cv2.warpAffine(image, M, (w, h))
cv2.imshow("Rotated by -90 degree", rotated)
cv2.waitKey(0)
rotated = imutils.rotate(image, 180)
cv2.imshow("Rotated by 180 degrees", rotated)
cv2.waitKey(0)
args = vars(ap.parse_args())
server_ip = args['server_ip']
rotation = float(args['rotate'])
# get the host name, initialize the video stream, and allow the
# camera sensor to warmup
rpiName = socket.gethostname()
video_stream = VideoStream(usePiCamera=True).start()
async_image_sender1 = AsyncImageSender(server_name=rpiName, server_ip=server_ip, port=5555, send_timeout=10, recv_timeout=10, show_frame_rate=10)
image_count = 0
print("Press ctrl-c to stop image sending")
sleep_time = 0.25
while True:
frame = video_stream.read()
if frame is not None:
if rotation != 0:
frame = imutils.rotate(frame, rotation)
async_image_sender1.send_frame_immediate(frame)
image_count += 1
time.sleep(sleep_time)
sumAreaDetect = [0, 0, 0]
pointMove = []
lencnts = 0
pathToDB = "/home/pi/FaceRecognizer/sorted_output_images"
#train
#igen_model, people= faceRec.train_model(pathToDB)
last_20 = [0 for i in range(20)]
final_5 = []
box_text = "Subject: "
while True:
# Get the next frame.
if args["video"] is None:
frame = vs.read()
frame = imutils.rotate(frame, 180)
else:
_, frame = vs.read()
text = "Unoccupied"
# If using a webcam instead of the Pi Camera,
# we take the extra step to change frame size.
if not usingPiCamera:
frame = imutils.resize(frame, width=frameSize[0])
if (frame is None):
print("not connect camera")
break
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
grayFace = cv2.equalizeHist(gray)
"""
faces = faceRec.detect_faces(grayFace)
# do Adrian Rosebrock
resized_easy = imutils.resize(image, height=200)
cv2.imshow("Redimensionamento com imutils", resized_easy)
cv2.waitKey(0)
# rotacionar uma imagem em 45 graus (sentido horário), calculando o centro da imagem,
# construir uma matriz de rotação e finalmente aplicação o "warp affine"
center = (w // 2, h // 2)
rotation_matrix = cv2.getRotationMatrix2D(center, -45, 1.0)
rotated = cv2.warpAffine(image, rotation_matrix, (w, h))
cv2.imshow("Rotação da imagem, com Rotation Matrix e Warp Affine", rotated)
cv2.waitKey(0)
# rotacionar uma imagem em 30 graus (sentido horário), usando a biblioteca imutils
# do Adrian Rosebrock
rotated_easy = imutils.rotate(image, -30)
cv2.imshow("Rotação da imagem, com imutils", rotated_easy)
cv2.waitKey(0)
# rotacionar uma imagem em 45 graus (sentido horário), usando a biblioteca imutils
# do Adrian Rosebrock, sem o cliping da mesma
# Nesta função, o nÃvel da rotação (em graus) deve ser o inverso (negativos - sentido antihorario,
# positivos - sentido horário)
rotated_bound = imutils.rotate_bound(image, 30)
cv2.imshow("Rotação da imagem, com imutils, sem clipping", rotated_bound)
cv2.waitKey(0)
# aplicar Gaussian Blur na imagem, com um kernel de 11x11 para suavizar,
# útil para reduzir "high frequency noise"
blurred = cv2.GaussianBlur(image, (11, 11), 0)
cv2.imshow("Borrada", blurred)
#from matplotlib import pyplot as plt
# Running function parameters
ap = argparse.ArgumentParser()
ap.add_argument('-i', '--image', required=True) #lokasi gambar
ap.add_argument('-j', '--parameter', required=True) #Jenis parameter
ap.add_argument('-k', '--node', required=True) #nomor node
args = vars(ap.parse_args())
eggparam = config[args["parameter"]][args["node"]]
eggparam = eggparam.split(',')
xcrop = int(eggparam[0])
ycrop = int(eggparam[1])
# Load, rotate and crop image
img = cv2.imread(args["image"])
img = imutils.rotate(img, 180)
h = img.shape[0]
w = img.shape[1]
img = img[0:h - ycrop, xcrop:w]
# Convert image to HSV and Split to get value part
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
hue, saturation, value = cv2.split(hsv)
# Apply adaptive threshold with Gaussian
thresh = cv2.adaptiveThreshold(value, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
cv2.THRESH_BINARY, 255, 17)
thresh = cv2.bitwise_not(thresh)
# Detecting blobs as noise
# Maximum blobs size
def read_digit(image_np,count,prev_reading):
height = image_np.shape[1]
width = image_np.shape[0]
gray = cv2.cvtColor(image_np,cv2.COLOR_BGR2GRAY)
plt.imshow(gray)
plt.show()
rot = imutils.rotate(gray,-90)
plt.imshow(rot)
plt.show()
thresh = cv2.adaptiveThreshold(rot.copy(),255,cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
cv2.THRESH_BINARY_INV,31,13)
kernel = cv2.getStructuringElement(cv2.MORPH_CROSS,(3,3))
thresh = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
# plt.imshow(thresh)
# plt.show()
im2, contours, hierarchy = cv2.findContours(thresh, 1, 2)
cnt = max(contours, key = cv2.contourArea)
#cnt = contours[0]
M = cv2.moments(cnt)
cimg = np.zeros_like(thresh)
cv2.drawContours(cimg, contours, contours.index(cnt), color=255, thickness=-1)
pts = np.where(cimg == 255)
plt.imshow(cimg)
plt.show()
arr = zip(pts[0],pts[1])
#print M
# cx = int(M['m10']/M['m00'])
# cy = int(M['m01']/M['m00'])
cx = width/2
cy = height /2
print('center: x %d y %d' %(cx,cy))
cv2.circle(rot,(cx,cy), 1, (0,0,255), -1)
max_dist = 0
max_pixel = None
for pixel in arr:
dist = math.sqrt(((cx - pixel[0])**2) + ((cy - pixel[1])**2))
if dist > max_dist:
max_pixel = pixel
max_dist = dist
print("max pixel: x %d y %d" %(max_pixel[1],max_pixel[0]))
cv2.circle(rot,(max_pixel[1],max_pixel[0]), 1, (0,0,255), -1)
plt.imshow(rot)
plt.show()
tx = max_pixel[1]
ty = max_pixel[0]
theta = math.atan2((ty-cy),(tx-cx))
phi = math.asin((cy - ty) / math.sqrt(((cy - ty)**2) + ((cx - tx) **2)))
degrees = math.degrees(theta)
if degrees >90 and degrees <=180:
angle = (degrees -180) - 90
else:
angle = degrees + 90
#print(theta)
print("degress %f " %(math.degrees(theta)))
print("actual angle %f" %(angle))
#print(phi)
angle = angle % 360
print("circle angle %f" %(angle))
angle = float(angle)
if angle>=0 and angle <= 36:
if count % 2 == 0:
reading = 0
else:
reading = 9
elif angle > 36 and angle <= 72:
if count % 2 == 0:
reading = 1
else:
reading = 8
elif angle > 72 and angle <=108:
if count % 2 == 0:
reading = 2
else:
reading = 7
elif angle > 108 and angle <= 144:
if count % 2 == 0:
reading = 3
else:
reading = 6
elif angle > 144 and angle <= 180:
if count % 2 == 0:
reading = 4
else:
reading = 5
elif angle > 180 and angle <=216:
if count % 2 == 0:
reading = 5
else:
reading = 4
elif angle > 216 and angle <=252:
if count % 2 == 0:
reading = 6
else:
reading = 3
elif angle > 252 and angle <= 288:
if count % 2 == 0:
reading = 7
else:
reading = 2
elif angle > 288 and angle <= 324:
if count % 2 == 0:
reading = 8
else:
reading = 1
elif angle > 324 and angle < 360:
if count % 2 == 0:
reading = 9
else:
reading = 0
# edge cases where the dial is close to a number
# must take into account previous reading to adjust for accuracy
if(prev_reading!=-1):
if((angle > 350 and angle<360) or angle < 10):
if (count %2 ==0):
if(prev_reading >= 5):
reading = 9
else:
reading = 0
else:
if(prev_reading >=5):
reading = 9
else:
reading = 0
elif(angle > 26 and angle <=46):
if count %2 ==0:
if(prev_reading >= 5):
reading = 0
else:
reading = 1
else:
if(prev_reading >=5):
reading = 8
else:
reading = 9
elif(angle > 62 and angle < 82):
if count %2 ==0:
if(prev_reading >= 5):
reading = 1
else:
reading = 2
else:
if(prev_reading >=5):
reading = 7
else:
reading = 8
elif(angle > 98 and angle < 118):
if count %2 ==0:
if(prev_reading >= 5):
reading = 2
else:
reading = 3
else:
if(prev_reading >=5):
reading = 6
else:
reading = 7
elif(angle > 134 and angle < 154):
if count %2 ==0:
if(prev_reading >= 5):
reading = 3
else:
reading = 4
else:
if(prev_reading >=5):
reading = 5
else:
reading = 6
elif(angle > 170 and angle < 190):
if count %2 ==0:
if(prev_reading >= 5):
reading = 4
else:
reading = 5
else:
if(prev_reading >=5):
reading = 4
else:
reading = 5
elif(angle > 206 and angle < 226):
if count %2 ==0:
if(prev_reading >= 5):
reading = 5
else:
reading = 6
else:
if(prev_reading >= 5):
reading = 3
else:
reading = 4
elif(angle > 242 and angle < 262):
if count %2 ==0:
if(prev_reading >= 5):
reading = 6
else:
reading = 7
else:
if(prev_reading >= 5):
reading = 2
else:
reading = 3
elif(angle > 278 and angle < 298):
if count %2 ==0:
if(prev_reading >= 5):
reading = 7
else:
reading = 8
else:
if(prev_reading >=5):
reading = 1
else:
reading = 2
elif(angle > 314 and angle < 334):
if count %2 ==0:
if(prev_reading >= 5):
reading = 8
else:
reading = 9
else:
if(prev_reading >= 5):
reading = 0
else:
reading = 1
print(reading)
return reading
ap.add_argument("-i", "--image", required=True, help="Path to image")
args = vars(ap.parse_args())
image = cv2.imread(args["image"])
cv2.imshow("Original", image)
cv2.waitKey(0)
shifted = imutils.translate(image, 25, 50)
# cv2.imshow('Shifted Down and Right', shifted)
# cv2.waitKey(0)
shifted = imutils.translate(image, -25, -50)
# cv2.imshow('Shifted Up and Left', shifted)
# cv2.waitKey(0)
rotated = imutils.rotate(image, 45)
# cv2.imshow('Rotated 45 degree', rotated)
# cv2.waitKey(0)
resized = imutils.resize(image, height=150)
# cv2.imshow('Resized to height 150', resized)
# cv2.waitKey(0)
flipped = cv2.flip(image, 1)
# cv2.imshow('Flipped Horizontally', flipped)
# cv2.waitKey(0)
flipped = cv2.flip(image, 0)
# cv2.imshow('Flipped Vertically', flipped)
# cv2.waitKey(0)
def rotateflipImg(Rotation_Angle=Rotation_Angle, Flip_X=Flip_X, noise_level=noise_level, step=step): # do_rotateflipImg
# def rotateflipImg(Rotation_Angle,Flip_X,noise_level,step):
print 'doing rotation or flipping', Rotation_Angle, ',', Flip_X, ',st', step, \
',noise=', noise_level, ',do_perspectiveTransform_in=', do_perspectiveTransform_in
print 'removing tmp data'
for i in xrange(len(settings.LABELS)):
dir = settings.tmp + settings.LABELS[i]
for f in os.listdir(dir):
if re.search('/*.jpg', f) is not None:
os.remove(os.path.join(dir, f))
filelist = glob.glob(INPUT_PATCHES_PATH + '/*')
# print 'filelist,', filelist
for filename in filelist:
for label in settings.LABELS:
if string.find(filename + '/', label) != -1: # hy:return index if found, or -1 if not found
INPUT_PATH = INPUT_PATCHES_PATH + label
OUTPUT_PATH = DATA_PATH + label
if DEBUG:
print 'If error occurs, possibly because some old file name errors, e.g. contain _ in the end'
# hy: copy patches to tmp folders
#cmd = 'cp -r ' + INPUT_PATH + '*' + ' ' + tmp_PATH + label # hy: copy recursively
#os.system(cmd)
filelist_tmp = glob.glob(INPUT_PATCHES_PATH + label + '*')
#filelist_tmp = glob.glob(tmp_PATH + label + '*')
random.shuffle(filelist_tmp)
# hy:use fixed additional amount of data
# random_add = 1
# random_add = 7*len(settings.LABELS)
# hy: use proportional setting
# random_add = int(len(filelist_tmp) * 0.70)
filelist_add = filelist_tmp[0:random_add]
# print len(filelist_tmp)
# for file in filelist_tmp: #hy if all data should be used
for file in filelist_add: # hy if only part data should be used
file.split('/')[-1]
# print file.split('/')[-1]
original = cv2.imread(file)
Height, Width, Channel = original.shape
# print original.shape
### Rotation start ##############################################################
if Anti_clockwise == 1 and Rotation_Angle <> 0:
# Counter-Clockwise: Zooming in, rotating, cropping
# rotated_tmp = cv2.resize(original, (Width + 40, Height + 20), interpolation=cv2.INTER_LINEAR)
# rotated_tmp = imutils.rotate(rotated_tmp, angle=Rotation_Angle)
# rotated = rotated_tmp[10:Height + 10, 20:Width + 20]
# print rotated.shape
# rotated = imutils.resize(rotated, width=Width,
# height=Height)
rotated = imutils.rotate(original, angle=Rotation_Angle)
new_file = OUTPUT_PATH + os.path.basename(os.path.normpath(file))[:-4] + '_st' + str(
step) + '_rotatedCC_' + str(
Rotation_Angle) + ImageType # hy: split by .jpg, with '.' to avoid extra '.' in file name
cv2.imwrite(new_file, rotated)
if Clockwise == 1 and Rotation_Angle <> 0:
# Clockwise
rotated_tmp = cv2.resize(original, (Width + 40, Height + 20), interpolation=cv2.INTER_LINEAR)
rotated_tmp = imutils.rotate(rotated_tmp, angle=Rotation_Angle * -1)
rotated = rotated_tmp[10:Height + 10, 20:Width + 20]
# print rotated.shape
rotated = imutils.resize(rotated, width=Width, height=Height)
new_file = OUTPUT_PATH + os.path.basename(os.path.normpath(file))[:-4] + '_st' + str(
step) + '_rotatedC_' + str(Rotation_Angle) + ImageType
cv2.imwrite(new_file, rotated)
### Rotation end ##############################################################
### Flipping begin ##############################################################
if Flip_X == 1:
flipped = cv2.flip(original, 0)
new_file = OUTPUT_PATH + os.path.basename(os.path.normpath(file))[:-4] + '_flippedX' + ImageType
cv2.imwrite(new_file, flipped)
if Flip_Y == 1:
flipped = cv2.flip(original, 1)
new_file = OUTPUT_PATH + os.path.basename(os.path.normpath(file))[:-4] + '_flippedY' + ImageType
cv2.imwrite(new_file, flipped)
if Flip_XY == 1:
flipped = cv2.flip(original, -1)
new_file = OUTPUT_PATH + os.path.basename(os.path.normpath(file))[:-4] + '_flippedXY' + ImageType
cv2.imwrite(new_file, flipped)
### Flipping end ##############################################################
### Perspective Transform begin #################################################
f_outs = []
if do_perspectiveTransform_in == 1:
# print 'file',file
# h, w, ch = original.shape
# print 'h,w,ch', h, w, ch
# rand1 = randint(2,30)
# rand2 = randint(2,30)
aspect_w = int(Aspect_Factor * Width)
for i in xrange(aspect_w, aspect_w + 2):
for j in xrange(14, 16):
pts1 = np.float32([[i, 0], [Width - i, 0], [j, Height - j], [Width - i, Height - j]])
# pts1 = np.float32([[rand1, 0], [patch_size+rand1, 0], [rand2, patch_size], [patch_size, patch_size]])
pts2 = np.float32([[0, 0], [Width, 0], [0, Height], [Width, Height]]) # leftT,rT,leftB,rB
# pts2 = np.float32([[0, 0], [patch_size, 0], [0, patch_size], [patch_size, patch_size]]) # leftT,rT,leftB,rB
M = cv2.getPerspectiveTransform(pts1, pts2)
dst = cv2.warpPerspective(original, M, (Width, Height)) # (w,h)
# dst = cv2.warpPerspective(img,M,(patch_size,patch_size)) #(w,h)
f_out = OUTPUT_PATH + os.path.basename(os.path.normpath(file))[:-4] + '_persp_' + str(i) + '_' + str(
j) + ImageType
# f_out = '../Data/data_1_mix/save/prep/' + os.path.basename(file) + '_persp_' + str(i) + '_' + str(j) + ImageType
print f_out
f_outs.append(f_out)
cv2.imwrite(f_out, dst)
print 'can generate num of new files with perspective transformation', len(f_outs)
### Perspective Transform end #################################################
### add noise begin ##############################################################
if noise_level <> 0:
img_gray = cv2.cvtColor(original, cv2.COLOR_BGR2GRAY)
# img = img.astype(np.float32)
img_noised = img_gray + np.random.rand(Width, Height) * noise_level
img_noised = (img_noised / np.max(img_noised)) * 255
new_file = OUTPUT_PATH + os.path.basename(os.path.normpath(file))[:-4] + '_st' + str(
step) + '_NOI' + ImageType
cv2.imwrite(new_file, img_noised)
# translate the image x-50 pixels to the left and y=100 pixels down
translated = imutils.translate(workspace, -50, 100)
cv2.imshow("Translated", translated)
cv2.waitKey(0)
# translate the image x=25 pixels to the right and y=75 pixels up
translated = imutils.translate(workspace, 25, -75)
cv2.imshow("Translated", translated)
cv2.waitKey(0)
cv2.destroyAllWindows()
# 2. ROTATION
# loop over the angles to rotate the image
for angle in range(0, 360, 90):
# rotate the image and display it
rotated = imutils.rotate(bridge, angle=angle)
cv2.imshow("Angle=%d" % (angle), rotated)
# wait for a keypress, then close all the windows
cv2.waitKey(0)
cv2.destroyAllWindows()
# 3. RESIZING
# loop over varying widths to resize the image to
for width in (400, 300, 200, 100):
# resize the image and display it
resized = imutils.resize(workspace, width=width)
cv2.imshow("Width=%dpx" % (width), resized)
# wait for a keypress, then close all the windows
cv2.waitKey(0)
def rotate(image, x):
#image = imutils.rotate_bound(image, -1) # not cutting information off
image = imutils.rotate(image, -x)
return image
#frame = cv2.imread(sys.argv[1])
# load the games image
#image = cv2.imread("tower.jpeg")
camera = cv2.VideoCapture(0)
#loop
while True:
#grabs current frame
(grabbed, frame) = camera.read()
#resizes video
frame = imutils.resize(frame, width = 300)
frame = imutils.rotate(frame, 180)
# find the red color game in the image
upper = [200,250,0]
lower = [20,180,0]
upper = np.array(upper,dtype = "uint8")
lower = np.array(lower,dtype = "uint8")
mask = cv2.inRange(frame, lower, upper)
#find contours in the masked frame and keep the largest one
(_, cnts, _) = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
leftmost = (0, 0)
rightmost = (0, 0)
topmost = (0, 0)
bottommost = (0, 0)
def OCR_Analyser(path, imageno):
imageno = ""
image = cv2.imread(path)
#Resigning the image to below dim as it has been found best during training.
dim = (348, 129)
image = cv2.resize(image, dim, interpolation=cv2.INTER_AREA)
imageoriginal = image
image = get_grayscale(image)
avgthresh = findavg(image)
cv2.imshow('1-GrayScale:' + str(imageno), image)
cv2.waitKey(0)
imagethresh = thresholding(image, avgthresh)
imagethresh = opening(imagethresh)
cv2.imshow('2-Threshold:' + str(imageno), imagethresh)
cv2.waitKey(0)
c = image
contours, hierarchy = cv2.findContours(imagethresh.copy(), cv2.RETR_TREE,
cv2.CHAIN_APPROX_SIMPLE)
val = -1
count = 0
for i in contours:
if len(i) == 4:
val = count
break
count = count + 1
if len(contours) == 0:
#print("###############################")
return "", imagethresh
c = max(contours, key=cv2.contourArea)
cv2.drawContours(imageoriginal, c, -1, (0, 255, 0), 3)
cv2.imshow('3-Contours:' + str(imageno), imageoriginal)
cv2.waitKey(0)
rect = cv2.minAreaRect(c)
#Finds angle and rotates the image so that its horizontal.
angle = rect[-1]
if abs(angle) < 90 - abs(angle):
angle = angle
else:
if angle < 0:
angle = 90 - abs(angle)
else:
angle = abs(angle) - 90
##%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
imageproc = imutils.rotate(imagethresh, angle)
contours, hierarchy = cv2.findContours(imageproc.copy(), cv2.RETR_TREE,
cv2.CHAIN_APPROX_SIMPLE)
c = max(contours, key=cv2.contourArea)
x, y, w, h = cv2.boundingRect(c)
cv2.drawContours(imageoriginal, c, -1, (0, 255, 0), 3)
imageproc = imageproc[y:y + h, x:x + w]
imageproc = erode(imageproc)
#imageproc = cv2.medianBlur(imageproc,3)
#imageproc=erode(imageproc)
#imageproc=dilate(imageproc)
#After Analysis I found pytesseract works best with dim= (385,90)
imageproc = cv2.resize(imageproc, (385, 90), interpolation=cv2.INTER_AREA)
#print("previous:",imageproc)
for i in range(11):
for j in range(384):
imageproc[i][j] = 255
imageproc[89 - i][j] = 255
for i in range(22):
for k in range(89):
imageproc[k][i] = 255
imageproc[k][384 - i] = 255
#Pytesseract configuration
oem = 0 #Legacy system
psm = 7 #Treats image as single text line
custom_config = r'--oem ' + str(oem) + ' --psm ' + str(
psm
) + '-c tessedit_char_whitelist=012345679abcdefghijklmnopqrstuvwlyz tessedit_char_blacklist=.-!/'
value = pytesseract.image_to_string(imageproc,
lang='eng',
config=custom_config)
value = value.encode('ascii', 'ignore')
cv2.imshow('4 - OCRImage:' + str(imageno), imageproc)
cv2.waitKey(0)
value = value.split()
finalresult = []
if len(value) > 2:
strval = " "
for i in value:
strval = strval + str(i) + " "
finalresult = strval
else:
firstword = firstwordpostprocessing(value)
secondword = ""
if len(value) > 0 and len(value) > 1:
secondword = secondwordpostprocessing(value)
#print("Firstword:",firstword)
#print("secondword:",secondword)
#print("Type of firstword:",type(str(firstword)))
#print("Type of secondword:",type(secondword))
finalresult = str(firstword) + " " + str(secondword)
return finalresult, imageproc
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required = True, help = "Path to the image")
args = vars(ap.parse_args())
image = cv2.imread(args["image"])
cv2.imshow("Original", image)
# (h, w) = image.shape[:2] # <~ clever way to get the height and width!
# center = (w // 2, h // 2)
# m1 = cv2.getRotationMatrix2D(center, 45, 1.0)
# rotated_up = cv2.warpAffine(image, m1, (w, h))
# cv2.imshow("Rotated by 45 Degrees", rotated_up)
# cv2.waitKey(0)
# m2 = cv2.getRotationMatrix2D(center, -90, 1.0)
# rotated_down = cv2.warpAffine(image, m2, (w, h))
# cv2.imshow("Rotated by -90 Degrees", rotated_down)
# cv2.waitKey(0)
rotated_up = imutils.rotate(image, 45, None, 1.0)
cv2.imshow("Rotated by 45 Degrees", rotated_up)
cv2.waitKey(0)
rotated_down = imutils.rotate(image, -90, None, 1.0)
cv2.imshow("Rotated by -90 Degrees", rotated_down)
cv2.waitKey(0)
cv2.waitKey(0)
import numpy as np
import argparse
import imutils
import cv2
ap=argparse.ArgumentParser()
ap.add_argument("-i","--image",required=True,help ="Path to the Image")
args=vars(ap.parse_args())
image = cv2.imread(args["image"])
cv2.imshow("Original",image)
rotated = imutils.rotate(image,45,center=(50,40),scale=2.0)
cv2.imshow("rotated", rotated)
cv2.waitKey(0)
# Construct the argument parser and parse the arguments
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required = True,
help = "Path to the image")
args = vars(ap.parse_args())
# Load the image and show it
image = cv2.imread(args["image"])
cv2.imshow("Original", image)
# Grab the dimensions of the image and calculate the center
# of the image
(h, w) = image.shape[:2]
center = (w / 2, h / 2)
# Rotate our image by 45 degrees
M = cv2.getRotationMatrix2D(center, 45, 1.0)
rotated = cv2.warpAffine(image, M, (w, h))
cv2.imshow("Rotated by 45 Degrees", rotated)
# Rotate our image by -90 degrees
M = cv2.getRotationMatrix2D(center, -90, 1.0)
rotated = cv2.warpAffine(image, M, (w, h))
cv2.imshow("Rotated by -90 Degrees", rotated)
# Finally, let's use our helper function in imutils.py to
# rotate the image by 180 degrees (flipping it upside down)
rotated = imutils.rotate(image, 180)
cv2.imshow("Rotated by 180 Degrees", rotated)
cv2.waitKey(0)
import cv2
import numpy as np
import matplotlib.pyplot as plt
import imutils
import math
width = 300
height = 300
filename = 'analog.jpeg'
org = cv2.imread(filename)
gray = cv2.cvtColor(org,cv2.COLOR_BGR2GRAY)
#gray = cv2.GaussianBlur(gray, (9, 9), 2,2)
img = imutils.resize(gray,width=300,height=300)
rot = imutils.rotate(img,-90)
# detect circles in the image
# circles = cv2.HoughCircles(img, cv2.HOUGH_GRADIENT, param1=100,param2=100)
# # ensure at least some circles were found
# if circles is not None:
# # convert the (x, y) coordinates and radius of the circles to integers
# circles = np.round(circles[0, :]).astype("int")
# # loop over the (x, y) coordinates and radius of the circles
# for (x, y, r) in circles:
# # draw the circle in the output image, then draw a rectangle
# # corresponding to the center of the circle
def read_images_online(filelist='',random_read=True,Anti_clockwise=0,Clockwise=0,Rotation_Angle=0,Flip_X=0,Flip_Y=0,noise_level=0,step=0):
print filelist,random_read,Anti_clockwise,Clockwise,Rotation_Angle,Flip_X,Flip_Y,noise_level,step
lable_file = open(filelist,'r')
lines = lable_file.readlines()
if random_read == True:
# make the file order in random way, this makes training more efficiently
random.shuffle(lines)
print 'loading images in random order'
else:
print 'loading sorted images'
images=[]
labels=[]
files=[]
label_sum=0
for item in lines:
filename,label = item.split()
#print filename
im = cv2.imread(filename)
im = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
#im = imutils.resize(im, width = 72, height = 72) # w=146, h=121
im = imutils.resize(im, width = settings.w_resize, height = settings.h_resize) # w=146, h=121
(settings.h_resize, settings.w_resize) = im.shape #hy:get suitable value for width
im=np.asarray(im, np.float32)
images.append(im)
labels.append(label)
files.append(filename)
############ add online data begin
# Rotation_Angle = randint(15, 170)
# noise_level = 0.01 * randint(1, 2)
### Rotation start ##############################################################
if Anti_clockwise == 1 and Rotation_Angle <> 0:
rotated = imutils.rotate(im, angle=Rotation_Angle)
images.append(rotated)
labels.append(label)
files.append(filename)
if Clockwise == 1 and Rotation_Angle <> 0:
# Clockwise
rotated_tmp = cv2.resize(im, (settings.w_resize + 40, settings.h_resize + 20), interpolation=cv2.INTER_LINEAR)
rotated_tmp = imutils.rotate(rotated_tmp, angle=Rotation_Angle * -1)
rotated = rotated_tmp[10:settings.h_resize + 10, 20:settings.w_resize + 20]
# print rotated.shape
rotated = imutils.resize(rotated, width=settings.w_resize, height=settings.h_resize)
images.append(rotated)
labels.append(label)
files.append(filename)
### Rotation end ##############################################################
### Flipping begin ##############################################################
if Flip_X == 1:
flipped = cv2.flip(im, 0)
images.append(flipped)
labels.append(label)
files.append(filename)
if Flip_Y == 1:
flipped = cv2.flip(im, 1)
images.append(flipped)
labels.append(label)
files.append(filename)
### Flipping end ##############################################################
### add noise begin ##############################################################
if noise_level <> 0:
img_gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
# img = img.astype(np.float32)
img_noised = img_gray + np.random.rand(settings.w_resize, settings.h_resize) * noise_level
img_noised = (img_noised / np.max(img_noised)) * 255
### add noise end ##############################################################
############ add online data end
label_sum += int(label)
images_np = np.asarray(images)
labels_np = np.asarray(labels)
return (images_np, labels_np, files)
cv2.imshow("Binary", img_out)
if pre[3] in PRE:
img_out = extract_contours(img_out)
cv2.imshow("Remove Contour", img_out)
cv2.waitKey(20000)
cv2.destroyAllWindows()
angle = 361
if MODE == modes[0]: # projection
angle = projection(img_out)
elif MODE == modes[1]: # hough
angle = hough(img_out)
rotated = imutils.rotate(img_orig, angle)
cv2.imshow("Rotated", rotated)
text = pytesseract.image_to_string(rotated)
print("Texto apos da rotação:")
print(text)
if angle >= 0:
print("Inclinação de %d° no sentido horário" % angle)
else:
print("Inclinação de %d° no sentido anti-horário" % (-angle))
if OUTPUT:
cv2.imwrite(OUTPUT, rotated)
cv2.waitKey(0)
import numpy as np
import imutils
import cv2
# Load an color image in grayscale
#img = cv2.imread('/home/hamed/WheelChair_Detect_copy/Data/Kiwa/2015-12-10_10.49_21.1.cam_81_1_020220.jpg')
img = cv2.imread('/home/hamed/Documents/Lego_copy/Data/hinten/20160421_133435_croppad_227_1_1.jpg')
cv2.imshow("test",img)
rotated=imutils.rotate(img,angle=5)
cv2.imshow("Rotated",rotated)
gray=cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
cv2.imshow("Gray",gray)
grayEq=gray
cv2.equalizeHist(gray, grayEq)
cv2.imshow("GrayEq", grayEq)
gray=cv2.flip(grayEq,-1)
cv2.imshow("flipped",gray)
cv2.imshow("reesize",imutils.resize(imutils.resize(gray,width=84),width=210)) #hy:updated size as per the example image old value 84,210
cv2.waitKey(0)
if key_input[pygame.K_ESCAPE]:
salir()
pygame.display.update()
fpsclock.tick(fps)
log()
# -- 2. Read the video stream
cap = cv.VideoCapture(0)
if not cap.isOpened:
print('--(!)Error opening video capture')
exit(0)
while True:
ret, frame = cap.read()
if frame is None:
print('--(!) No captured frame -- Break!')
break
fps = cap.get(5)
cv.putText(frame, str(fps), (50, 50), cv.FONT_HERSHEY_SIMPLEX, 1,
(0, 255, 255))
frame = imutils.rotate(frame, 180 - pos_pan)
cv.imshow('Capture - Rotate on pan', frame)
if cv.waitKey(1) == ord('q'):
break
mover_pygame()
def detect_landmarks(img):
import face_alignment
import imutils
import numpy as np
import dlib
import cv2
fa = face_alignment.FaceAlignment(face_alignment.LandmarksType._2D,
flip_input=False)
p = "shape_predictor_68_face_landmarks.dat"
detector = dlib.get_frontal_face_detector()
predictor = dlib.shape_predictor(p)
res = cv2.resize(img, dsize=(80, 80), interpolation=cv2.INTER_CUBIC)
landmarks = np.empty([])
rects = detector(res, 0)
if (len(rects)) != 0:
rect = rects[0]
landmarks = predictor(res, rect)
landmarks = process_landmarks(landmarks)
landmarks = 0.55 * np.array(landmarks)
else:
rotated = imutils.rotate(res, 90)
rects = detector(rotated, 0)
if (len(rects)) != 0:
rect = rects[0]
landmarks = predictor(rotated, rect)
landmarks = process_landmarks(landmarks)
landmarks = 0.55 * np.array(landmarks)
else:
rotated = imutils.rotate(res, -90)
rects = detector(rotated, 0)
if (len(rects)) != 0:
rect = rects[0]
landmarks = predictor(rotated, rect)
landmarks = process_landmarks(landmarks)
landmarks = 0.55 * np.array(landmarks)
else:
preds = fa.get_landmarks(img)
if preds is not None:
points = preds[0]
potints = np.array(points)
landmarks = []
xl = []
yl = []
for i in range(0, 68):
xl.append(points[i, 0])
yl.append(points[i, 1])
xmean = np.mean(xl)
ymean = np.mean(yl)
xcentral = [(x - xmean) for x in xl]
ycentral = [(y - ymean) for y in yl]
for j in range(0, 68):
for k in range(j + 1, 68):
if j != k:
distx = (xcentral[j] - xcentral[k])
disty = (ycentral[j] - ycentral[k])
landmarks.append(distx)
landmarks.append(disty)
landmarks = np.array(landmarks)
else:
preds = fa.get_landmarks(res)
if preds is not None:
points = preds[0]
potints = np.array(points)
landmarks = []
xl = []
yl = []
for i in range(0, 68):
xl.append(points[i, 0])
yl.append(points[i, 1])
xmean = np.mean(xl)
ymean = np.mean(yl)
xcentral = [(x - xmean) for x in xl]
ycentral = [(y - ymean) for y in yl]
for j in range(0, 68):
for k in range(j + 1, 68):
if j != k:
distx = (xcentral[j] - xcentral[k])
disty = (ycentral[j] - ycentral[k])
landmarks.append(distx)
landmarks.append(disty)
landmarks = 0.55 * np.array(landmarks)
#print('Face align size: '+str(landmarks.size))
return landmarks
def face_recognition():
while not (GPIO.input(11) or GPIO.input(13) or GPIO.input(15)):
#while True:
print("yes3")
cam.capture("/home/pi/Desktop/switch/face1.png")
#img = np.asarray(Image.open("/home/pi/Desktop/switch/FacialRecognitionProject/face1.png"))
#img =cam.read()
img = cv2.imread("/home/pi/Desktop/switch/face1.png")
img = cv2.flip(img, -1) # Flip vertically
img = imutils.rotate(img, angle=180)
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
faces = faceCascade.detectMultiScale(
gray,
scaleFactor=1.2,
minNeighbors=5,
minSize=(int(minW), int(minH)),
)
for (x, y, w, h) in faces:
cv2.rectangle(img, (x, y), (x + w, y + h), (0, 255, 0), 2)
r = (2 * x + w) / 2
id, confidence = recognizer.predict(gray[y:y + h, x:x + w])
# Check if confidence is less them 100 ==> "0" is perfect match
if ((100 - confidence) > 40):
id = names[id]
confidence = " {0}%".format(round(100 - confidence))
if (r > 540):
engine = pyttsx3.init()
engine.say("go left")
engine.say("it's " + str(id))
engine.runAndWait()
else:
engine = pyttsx3.init()
engine.say("go right")
engine.say("it's " + str(id))
engine.runAndWait()
#file1 = open("MyFile.txt","a")
#file1.write("\n"+str(id)+ " at: "+str(datetime.datetime.now()))
else:
id = "unknown"
confidence = " {0}%".format(round(100 - confidence))
if (r > 540):
engine = pyttsx3.init()
engine.say("go left")
engine.runAndWait()
else:
engine = pyttsx3.init()
engine.say("go right")
engine.runAndWait()
cv2.putText(img, str(id), (x + 5, y - 5), font, 1,
(255, 255, 255), 2)
cv2.putText(img, str(confidence), (x + 5, y + h - 5), font, 1,
(255, 255, 0), 1)
cv2.imshow('camera', img)
print("yes3")
#if str(id)!= "unknown":
# speechCallName(str(id))
k = cv2.waitKey(10) & 0xff # Press 'ESC' for exiting video
if k == 27:
break
import numpy as np
import argparse
import imutils
import cv2
ap = argparse.ArgumentParser()
ap.add_argument('-i', '--image', required=True, help="Path")
args = vars(ap.parse_args())
image = cv2.imread(args['image'])
cv2.imshow("Oriji", image)
(h, w) = image.shape[:2]
centre = (w // 2, h // 2)
M = cv2.getRotationMatrix2D(centre, 45, 1.0)
rotated = cv2.warpAffine(image, M, (w, h))
cv2.imshow("45", rotated)
M = cv2.getRotationMatrix2D(centre, -90, 1.0)
rotated = cv2.warpAffine(image, M, (w, h))
cv2.imshow("-90", rotated)
image = imutils.rotate(image, centre, 200, 2, w, h)
cv2.imshow("270", image)
cv2.waitKey(0)
while cap.isOpened():
ret,frame = cap.read()
if ret == True:
cv2.imshow('window-name',frame)
cv2.imwrite("./output/frame%d.jpg" % count, frame)
count = count + 1
if cv2.waitKey(10) & 0xFF == ord('q'):
break
else:
break
cap.release()
cv2.destroyAllWindows()
# car image -> grayscale image -> binary image
car_image = imread("./output/frame%d.jpg"%(count-1), as_gray=True)
car_image = imutils.rotate(car_image, 270)
# car_image = imread("car.png", as_gray=True)
# it should be a 2 dimensional array
print(car_image.shape)
'''
the next line is not compulsory however, a grey scale pixel
in skimage ranges between 0 & 1. multiplying it with 255
will make it range between 0 & 255 (something we can relate better with
'''
gray_car_image = car_image * 255
fig, (ax1, ax2) = plt.subplots(1, 2)
ax1.imshow(gray_car_image, cmap="gray")
threshold_value = threshold_otsu(gray_car_image)
binary_car_image = gray_car_image > threshold_value
import argparse
import cv2
import imutils
ap = argparse.ArgumentParser()
ap.add_argument('-i', '--image', required=True, help='Path to input image')
args = vars(ap.parse_args())
image = cv2.imread(args['image'])
#cv2.imshow('Image', image)
flipped = cv2.flip(image, 1)
cv2.imshow('Flipped', flipped)
print(flipped[235, 259])
rotated = imutils.rotate(flipped, 45)
flipped = cv2.flip(rotated, 0)
cv2.imshow('Flipped', flipped)
print(flipped[189, 441])
#flipped = cv2.flip(image, -1)
#cv2.imshow('F
# lipped', flipped)
cv2.waitKey(0)
def detect(request):
s=str(request.method)
fil=[]
print("request files",request.FILES)
for x in request.FILES:
fil.append(str(request.FILES[x]))
for y in fil:
from time import time
t1=time()
im=cv.imread(y)
#im=imutils.rotate(im,-90)
#im=im[im.shape[1]//2:,:]
obj,lst=object_detect(im)
tobj=obj
tobj=imutils.rotate(tobj,-90)
print("time for object detection",time()-t1)
img=lanedetect(tobj)
#img=imutils.rotate(img,90)
for l in lst:
l.assign_dist(calc_distance(l.middle_x,l.middle_y))
#l.assign_offset()
try:
nearest_obstacle=find_nearest_obstacle(lst)
except Exception as e:
print("In calling function")
print(e)
try:
cv2.line(obj,(int(pt[0]),int(pt[1])),(nearest_obstacle['middle_x'],nearest_obstacle['middle_y']),5)
except Exception as e:
print(e)
cv2.imshow("image",obj)
cv2.waitKey(0)
data = {"success": False,"hiii":"hello","method":s,"File":fil}
data['offset']=getOffsetStatus()
data['obstacles']=findObstaclesInPath(nearest_obstacle)
return JsonResponse(data)
plt.yticks([])
# ------------------------------------
a = fig.add_subplot(2, 3, 4)
imgplot = plt.imshow(imageInputOrig)
a.set_title('Before')
plt.xticks([])
plt.yticks([])
a = fig.add_subplot(2, 3, 5)
imgplot = plt.imshow(whiteImg)
a.set_title('Landmarks')
plt.xticks([])
plt.yticks([])
imageInput = imutils.rotate(imageInput, 45)
a = fig.add_subplot(2, 3, 6)
imgplot = plt.imshow(newImage)
a.set_title('After')
plt.xticks([])
plt.yticks([])
fname = "results/" + "result_" + args["image"][1]
plt.show()
'''
plt.imshow(imageInput)
limpiar();
print("#########################################################");
# Ej 016_16
print("Inicio ej016_16 - ");
center = (w // 2, h // 2)
M = cv2.getRotationMatrix2D(center, -45, 1.0)
rotated = cv2.warpAffine(imagen, M, (w, h))
cv2.imshow("Opencv2 Rotation", rotated)
cv2.waitKey(1500)
cv2.destroyAllWindows()
#print (input("Fin ej016_16 \n continuar?"));
limpiar();
print("#########################################################");
# Ej 016_17
print("Inicio ej016_17 - ");
rotated = imutils.rotate(imagen, 45)
cv2.imshow("Imutils Rotation", rotated)
cv2.waitKey(1500)
cv2.destroyAllWindows()
#print (input("Fin ej016_18 \n continuar?"));
limpiar();
print("#########################################################");
# Ej 016_18
print("Inicio ej016_18 - ");
blurred = cv2.GaussianBlur(imagen, (11, 11), 0)
cv2.imshow("Blurred", blurred)
cv2.waitKey(1500)
cv2.destroyAllWindows()
#print (input("Fin ej016_18 \n continuar?"));
limpiar();
print("#########################################################");
rotated = cv2.warpAffine(img, M, (w, h))
cv2.imshow("rotated", rotated)
#by 90 degree and scale 0.5
M = cv2.getRotationMatrix2D((cX, cY), 90, 0.5)
rotated2 = cv2.warpAffine(img, M, (w, h))
cv2.imshow("rotated2", rotated2)
#by -90 degree and scale 2.0 (double)
M = cv2.getRotationMatrix2D((cX, cY), -90, 2.0)
rotated3 = cv2.warpAffine(img, M, (w, h))
cv2.imshow("rotated3", rotated3)
#using image utils lib
rotated_utils = imutils.rotate(img, 180, None, 0.5)
cv2.imshow("rotated_utils", rotated_utils)
"""
def rotate(img,angle,center=None,scale=1.0):
h,w = img.shape[:2]
if center is None:
center = (w//2,h//2)
M = cv2.getRotationMatrix2D(center,angle,scale)
rotate = cv2.wrapAffine(img,M,(w,h))
return rotate
"""
import cv2
# Read the existing image.
image_packet = impb.Image()
try:
f = open("image_data.proto.bin", "rb")
image_packet.ParseFromString(f.read())
f.close()
except IOError:
print("image_data.proto.bin" + ": Could not open file.")
with open('picture_out.jpg', 'wb') as f:
f.write(image_packet.image_data)
image_transform = cv2.imread("picture_out.jpg")
(h, w, d) = image_transform.shape
print("width={}, height={}, depth={}".format(w, h, d))
(B, G, R) = image_transform[100, 50]
print("R={}, G={}, B={}".format(R, G, B))
roi = image_transform[60:160, 320:420]
resized = cv2.resize(image_transform, (100, 100))
rotated = imutils.rotate(image_transform, -45)
#with open('picture_out.jpg', 'wb') as imagefile:
#imagefile.write(bytearray(image_packet.SerializeToString()))
#bytesIOImage = io.BytesIO(bytearray(image_packet.SerializeToString()))
#bytesIOImage.seek(0)
#image = Image.open(bytesIOImage)
#image.save('picture_out.jpg')
nargs='+',
help="height and width to crop")
ap.add_argument("-r", "--ratio", required=True, help="ratio for shrinking")
args = ap.parse_args()
workingPath = args.srcPath
targetPath = args.tgtPath
ang = float(args.angle)
cp = args.cropPercent
r = float(args.ratio)
# print(float(cp[1]))
imageFiles = os.listdir(workingPath)
rgbIm = []
for im in imageFiles:
if im.find(".jpg") != -1:
rgbIm.append(im)
# Detect each individual image
for imf in rgbIm:
imgFile = cv2.imread(workingPath + "\\" + imf)
rotated = imutils.rotate(imgFile, ang)
imgCrop = rotated[(int(rotated.shape[0] *
(float(cp[0])))):(int(rotated.shape[0] *
(float(cp[1]))) - 1),
(int(rotated.shape[1] *
(float(cp[2])))):(int(rotated.shape[1] *
(float(cp[3]))) - 1)]
resizedImage = cv2.resize(
imgCrop, (int(imgCrop.shape[1] * r), int(imgCrop.shape[0] * r)),
interpolation=cv2.INTER_AREA)
print(targetPath + "\\" + imf)
cv2.imwrite(targetPath + "\\" + imf, resizedImage)
cv2.imshow("Imutils Resize", resized)
cv2.waitKey(0)
# let's rotate an image 45 degrees clockwise using OpenCV by first
# computing the image center, then constructing the rotation matrix,
# and then finally applying the affine warp
center = (w // 2, h // 2)
""" cv2.getRotationMatrix2D(center, angle, scale) """
M = cv2.getRotationMatrix2D(center, -45, 1.0)
""" cv2.warpAffine(src, M, dsize[, dst[, flags[, borderMode[, borderValue]]]]) """
rotated = cv2.warpAffine(image, M, (w, h))
cv2.imshow("OpenCV Rotation", rotated)
cv2.waitKey(0)
# rotation can also be easily accomplished via imutils with less code
rotated = imutils.rotate(image, -45)
cv2.imshow("Imutils Rotation", rotated)
cv2.waitKey(0)
# OpenCV doesn't "care" if our rotated image is clipped after rotation
# so we can instead use another imutils convenience function to help
# us out
rotated = imutils.rotate_bound(image, 45)
cv2.imshow("Imutils Bound Rotation", rotated)
cv2.waitKey(0)
# apply a Gaussian blur with a 11x11 kernel to the image to smooth it,
# useful when reducing high frequency noise
""" cv2.GaussianBlur(src, ksize, sigmaX[, dst[, sigmaY[, borderType]]])kSize es """
blurred = cv2.GaussianBlur(image, (11, 11), 0)
cv2.imshow("Blurred", blurred)
nW = int((w * cos) + (h * sin))
nH = int((w * sin) + (h * cos))
M[0, 2] += (nW / 2) - cX
M[1, 2] += (nH / 2) - cY
return cv2.warpAffine(image, M, (nW, nH))
ap = argparse.ArgumentParser()
ap.add_argument("-i", "--image", required=True, help="path to the image file")
args = vars(ap.parse_args())
image = cv2.imread(args["image"])
for angle in np.arange(0, 360, 15):
rotated = imutils.rotate(image, -angle)
cv2.imshow("Rotated (problematic)", rotated)
cv2.waitKey(500)
for angle in np.arange(0, 360, 15):
rotated = imutils.rotate_bound(image, angle)
cv2.imshow("Rotated (Correct)", rotated)
cv2.waitKey(500)
for angle in np.arange(0, 360, 15):
rotated = rotate_bound_mine(image, angle)
cv2.imshow("Rotated (Correct)", rotated)
cv2.waitKey(500)
def rotation_not_90_func(img, label, thetha):
rotated = imutils.rotate(img, thetha)
rotated_label = imutils.rotate(label, thetha)
return rotate_max_area(img, rotated, rotated_label, thetha)
def orientation():
rotation_ratio = 72
angles = 5
myList = [None] * rotation_ratio
mask1List = [None] * rotation_ratio
rotate1List = [None] * rotation_ratio
mask2List = [None] * rotation_ratio
rotate2List = [None] * rotation_ratio
blackPixList = [None] * rotation_ratio
differList = [None] * rotation_ratio
i = j = k = 0
lower = [(0, 100, 100)]
upper = [(100, 255, 255)]
img1 = cv.imread('images_temp/normal_crop_img.jpg')
img2 = cv.imread('images_temp/saved_img.jpg')
for mask2angle in np.arange(0, 360, angles):
rotate2List[int(j)] = imutils.rotate(img2, mask2angle)
edged2 = cv.Canny(rotate2List[j], 0, 200)
pts2 = np.argwhere(edged2 > 0)
y12, x12 = pts2.min(axis=0)
y22, x22 = pts2.max(axis=0)
mask2List[int(j)] = y22
j = j + 1
for mask1angle in np.arange(0, 360, angles):
rotate1List[int(k)] = imutils.rotate(img1, mask1angle)
edged1 = cv.Canny(rotate1List[k], 0, 200)
pts1 = np.argwhere(edged1 > 0)
y11, x11 = pts1.min(axis=0)
y21, x21 = pts1.max(axis=0)
mask1List[int(k)] = y21
k = k + 1
edged1 = cv.Canny(rotate1List[mask1List.index(min(mask1List))], 0, 200)
edged2 = cv.Canny(rotate2List[mask2List.index(min(mask2List))], 0,
200) #150 550 # 0 200
pts1 = np.argwhere(edged1 > 0)
pts2 = np.argwhere(edged2 > 0)
y11, x11 = pts1.min(axis=0)
y12, x12 = pts2.min(axis=0)
y21, x21 = pts1.max(axis=0)
y22, x22 = pts2.max(axis=0)
croppedimg1 = rotate1List[mask1List.index(min(mask1List))][y11:y21,
x11:x21]
croppedimg2 = rotate2List[mask2List.index(min(mask2List))][y12:y22,
x12:x22]
lower = np.array(lower, dtype="uint8")
upper = np.array(upper, dtype="uint8")
mask1 = cv.inRange(croppedimg1, lower, upper)
mask2 = cv.inRange(croppedimg2, lower, upper)
img_shape1 = croppedimg1.shape
img_shape2 = croppedimg2.shape
if img_shape1 == img_shape2:
crop_mask1 = mask1
crop_mask2 = mask2
if img_shape1[0] > img_shape2[0]:
if img_shape1[1] >= img_shape2[1]:
crop_mask1 = mask1[0:img_shape2[0], 0:img_shape2[1]]
crop_mask2 = mask2[0:img_shape2[0], 0:img_shape2[1]]
if img_shape1[1] < img_shape2[1]:
crop_mask1 = mask1[0:img_shape2[0], 0:img_shape1[1]]
crop_mask2 = mask2[0:img_shape2[0], 0:img_shape1[1]]
elif img_shape1[0] < img_shape2[0]:
if img_shape1[1] >= img_shape2[1]:
crop_mask1 = mask1[0:img_shape1[0], 0:img_shape2[1]]
crop_mask2 = mask2[0:img_shape1[0], 0:img_shape2[1]]
if img_shape1[1] < img_shape2[1]:
crop_mask1 = mask1[0:img_shape1[0], 0:img_shape1[1]]
crop_mask2 = mask2[0:img_shape1[0], 0:img_shape1[1]]
for angle in np.arange(0, 360, angles):
myList[int(i)] = imutils.rotate(crop_mask2, angle)
i = i + 1
for a in range(len(myList)):
differList[a] = cv.subtract(crop_mask1, myList[a])
for b in range(len(myList)):
blackPixList[b] = np.sum(differList[b] == 0)
orientationValue = ((((mask2List.index(min(mask2List))) * angles) +
((blackPixList.index(max(blackPixList))) * angles)) -
360) - ((mask1List.index(min(mask1List))) * angles)
if orientationValue <= (-360):
orientationValue = orientationValue + 360
if orientationValue >= 360:
orientationValue = orientationValue - 360
print("max ", max(blackPixList), "index ",
blackPixList.index(max(blackPixList)))
print("orientation :", orientationValue)
cv.imwrite('images_temp/mask1.jpg', crop_mask1)
cv.imwrite('images_temp/mask2.jpg', crop_mask2)
cv.imwrite('images_temp/mask_final.jpg',
myList[blackPixList.index(max(blackPixList))])
# to be detected in the HSV color space
colorLower = (24, 100, 100)
colorUpper = (44, 255, 255)
# Start with LED off
print("\n Starting..... ==> Press 'q' to quit Program \n")
GPIO.output(redLed, GPIO.LOW)
ledOn = False
# loop over the frames from the video stream
while True:
# grab the next frame from the video stream, Invert 180o, resize the
# frame, and convert it to the HSV color space
frame = vs.read()
frame = imutils.resize(frame, width=500)
frame = imutils.rotate(frame, angle=180)
hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
# construct a mask for the obect color, then perform
# a series of dilations and erosions to remove any small
# blobs left in the mask
mask = cv2.inRange(hsv, colorLower, colorUpper)
mask = cv2.erode(mask, None, iterations=2)
mask = cv2.dilate(mask, None, iterations=2)
# find contours in the mask and initialize the current
# (x, y) center of the object
cnts = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if imutils.is_cv2() else cnts[1]
center = None
count = 0
print("path:" + path)
# loop over the frames of the video
while True:
# grab the current frame and initialize the occupied/unoccupied
# text
frame = vs.read()
frame = frame if args.get("video", None) is None else frame[1]
text = "Emtpy"
# if the frame could not be grabbed, then we have reached the end
# of the video
if frame is None:
break
# resize the frame, convert it to grayscale, and blur it
frame = imutils.resize(frame, width=500)
frame = imutils.rotate(frame, 90)
# mask
mask = np.zeros(frame.shape, dtype=np.uint8)
roi_corners = np.array([[(300, 10), (490, 310), (10, 290), (210, 10)]],
dtype=np.int32)
# fill the ROI so it doesn't get wiped out when the mask is applied
channel_count = frame.shape[2] # i.e. 3 or 4 depending on your image
ignore_mask_color = (255, ) * channel_count
cv2.fillPoly(mask, roi_corners, ignore_mask_color)
# apply the mask
frame = cv2.bitwise_and(frame, mask)
hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
greenMask = cv2.inRange(hsv, (26, 10, 30), (255, 255, 255))
#Initial run camera
print("Staring the camera. Please wait for 2 seconds")
vs = VideoStream(0).start()
time.sleep(2.0)
# Joycon color
joyconLower = (99, 100, 100)
joyconUpper = (119, 255, 255)
while True:
sleep(1)
print("Grab Frame")
tmpFrame = vs.read()
tmpFrame = imutils.resize(tmpFrame, width=500)
tmpFrame = imutils.rotate(tmpFrame, angle=180)
hsv = cv2.cvtColor(tmpFrame, cv2.COLOR_BGR2HSV)
print("Create tmpMask")
tmpMask = cv2.inRange(hsv, joyconLower, joyconUpper)
tmpMask = cv2.erode(tmpMask, None, iterations=2)
tmpMask = cv2.dilate(tmpMask, None, iterations=2)
print("Find Joycon..")
countsCircle = cv2.findContours(tmpMask.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)
countsCircle = countsCircle[0] if imutils.is_cv4() else countsCircle[1]
center = None
if len(countsCircle) > 0:
print("Find Object")
# keep looping
while True:
# grab the current frame
(grabbed, frame) = camera.read()
# if we are viewing a video and we did not grab a frame,
# then we have reached the end of the video
if args.get("video") and not grabbed:
break
# resize the frame, blur it, and convert it to the HSV
# color space
frame = imutils.resize(frame, width=600)
if args.get("video"):
frame = imutils.rotate(frame, -90)
# blurred = cv2.GaussianBlur(frame, (11, 11), 0)
hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
# construct a mask for the color "green", then perform
# a series of dilations and erosions to remove any small
# blobs left in the mask
mask = cv2.inRange(hsv, greenLower, greenUpper)
mask = cv2.erode(mask, None, iterations=2)
mask = cv2.dilate(mask, None, iterations=2)
# find contours in the mask and initialize the current
# (x, y) center of the ball
cnts = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL,
cv2.CHAIN_APPROX_SIMPLE)[-2]
cnts = sorted(cnts, key=cv2.contourArea, reverse=True)[:balls]
