def dashboard():
if 'email' not in session:
return redirect(
url_for('login')) #requires an account to access this page
resume = True
photo = True
emotions = False
photo_feedback = False
resume_feedback = False
score = False
error = False
if request.method == 'POST':
user_info = users.find_one({'email': session['email']})
if 'resume' in request.form:
if user_info['resume'] != '':
resume_data = user_info['resume']
fh = open("sentiment_results.txt", 'w')
parsed_output = json.loads(resume_data)
fh.write(
json.dumps(parsed_output["DocumentElement"]["Resume"]
["Experience"],
indent=4,
sort_keys=True))
fh.close()
results = sentiment.analyze("sentiment_results.txt")
score = results[0]
resume_feedback = analyzeResume(score)
else:
resume = False
if 'photo' in request.form:
if fs.exists(user_info['image']):
image_data = fs.get(user_info['image'])
f = open(UPLOAD_FOLDER + user_photo, 'wb')
f.write(image_data.read())
f.close()
emotions = detect_face.detect_faces(UPLOAD_FOLDER + user_photo)
photo_feedback = analyzeRelease(
emotions["anger"], emotions["joy"], emotions["surprise"],
emotions["blurred"],
emotions["headwear"]) if emotions else False
error = False if photo_feedback else True
else:
photo = False
return render_template("dashboard.html",
emotions=emotions,
photo=photo,
score=score,
resume=resume,
photo_feedback=photo_feedback,
resume_feedback=resume_feedback,
error=error)
def extractface():
responseArray = {}
if request.method == 'POST':
images = request.files.to_dict()
for image in images: #image will be the key
print(images[image]) #this line will print value for the image key
file_name = images[image].filename
extension = os.path.splitext(file_name)[1]
f_name = str(uuid.uuid4()) + extension
images[image].save(os.path.join('./source', f_name))
#responseArray.append(detect_face.detect_faces(f_name))
responseArray[f_name] = detect_face.detect_faces('./source/' +
f_name)
return jsonify(responseArray)
else:
return jsonify(msg='error')
def vid2frames(path=datadir + "savee/AudioVisualClip/DC/a1.avi",
subject="DC",
vid_label="a1"):
# noinspection PyArgumentList
cap = cv2.VideoCapture(path)
# print(cap)
n = 0
while cap.isOpened():
# Capture frame-by-frame
ret, frame = cap.read()
# print(str(n) +"\r")
if ret is True:
# Display the resulting frame
# cv2.imshow('Frame', frame)
frame = get_largest_face(
frame,
detect_faces(
cv2.CascadeClassifier('lbpcascade_frontalface.xml'),
frame))
try:
frame = cv2.resize(frame, (48, 48),
interpolation=cv2.INTER_CUBIC)
except cv2.error as e:
frame = np.zeros((48, 48))
cv2.imwrite(
datadir + 'frames/' + subject + '_' + vid_label + '_' +
str(n).zfill(5) + '.png', frame)
n = n + 1
# print(n)
# Press Q on keyboard to exit
if cv2.waitKey(25) & 0xFF == ord('q'):
break
# Break the loop
else:
break
# When everything done, release the video capture object
cap.release()
# Closes all the frames
cv2.destroyAllWindows()
def show_webcam_and_run(model,
emoticons,
window_size=None,
window_name='webcam',
update_time=10):
"""
Shows webcam image, detects faces and its emotions in real time and draw emoticons over those faces.
:param model: Learnt emotion detection model.
:param emoticons: List of emotions images.
:param window_size: Size of webcam image window.
:param window_name: Name of webcam image window.
:param update_time: Image update time interval.
"""
cv2.namedWindow(window_name, WINDOW_NORMAL)
if window_size:
width, height = window_size
cv2.resizeWindow(window_name, width, height)
vc = cv2.VideoCapture(0)
if vc.isOpened():
read_value, webcam_image = vc.read()
else:
print("webcam not found")
return
while read_value:
for normalized_face, (x, y, w, h) in detect_faces(webcam_image):
prediction = model.predict(normalized_face) # do prediction
if cv2.__version__ != '3.1.0':
prediction = prediction[0]
image_to_draw = emoticons[prediction]
draw_with_alpha(webcam_image, image_to_draw, (x, y, w, h))
cv2.imshow(window_name, webcam_image)
read_value, webcam_image = vc.read()
key = cv2.waitKey(update_time)
if key == 27: # exit on ESC
break
cv2.destroyWindow(window_name)
def extract_faces(emotions):
"""
Crops faces in emotions images.
:param emotions: List of emotions names.
"""
print("Extracting faces")
for emotion in emotions:
photos = glob.glob('data/sorted_set/%s/*' % emotion)
for file_number, photo in enumerate(photos):
frame = cv2.imread(photo)
normalized_faces = detect_faces(frame)
os.remove(photo)
for face in normalized_faces:
try:
cv2.imwrite("data/sorted_set/%s/%s.png" % (emotion, file_number + 1), face[0]) # write image
except:
print("error in processing %s" % photo)
def video_capture(image_based=False):
cv2.namedWindow("exit on ESC")
# to capture video from cv2
# fourcc = cv2.VideoWriter_fourcc(*'XVID')
# out = cv2.VideoWriter('output.avi', fourcc, (640,
vc = cv2.VideoCapture(0)
frame = 0
video = []
temp_video = []
fill = np.zeros(shape)
if vc.isOpened(): # try to get the first frame
rval, frame = vc.read()
else:
rval = False
f = cv2.flip(frame, 1)
frames_passed = 0
frame_rate = 2
plot = frame[-120:, 0:120]
(angry, disgust, fear, happy, neutral, sad, surprise) = (0, 0, 0, 0, 0, 0,
0)
while rval:
cv2.imshow("exit on ESC", f)
rval, frame = vc.read()
# tilt optimization req
# temp = crop_rot_images(frame, lbp_face_cascade,draw_face=True)
# if frames_passed % frame_rate == 0:
temp = get_largest_face(
frame,
detect_faces(cv2.CascadeClassifier('lbpcascade_frontalface.xml'),
frame),
draw_face=True)
# temp = draw_faces(frame, detect_faces(lbp_face_cascade, frame))
if temp.shape != (0, 0, 3):
frame = cv2.flip(frame, 1)
if image_based:
angry, disgust, fear, happy, neutral, sad, surprise = predict_emotion_image(
temp)
else:
# implemented for only transfer learning part
if not frames_passed % frame_rate == 3:
temp = cv2.cvtColor(temp, cv2.COLOR_BGR2GRAY)
temp = cv2.resize(temp, (48, 48),
interpolation=cv2.INTER_AREA)
resized = (np.moveaxis(temp, -1, 0)).reshape(
(1, 1, 48, 48))
# print(resized.shape)
vector = intermediate_layer_model.predict(resized)
# print(vector.shape)
if len(video) >= 60:
video.append(vector.reshape(4608))
temp_video = video[1:61]
else:
video.append(vector.reshape(4608))
temp_video = np.concatenate((video, fill),
axis=0)[0:60]
# out.write(frame)
angry, disgust, fear, happy, neutral, sad, surprise = predict_emotion_video(
temp_video)
frames_passed = frames_passed + 1
print(
put_emoji(angry, disgust, fear, happy, neutral, sad, surprise))
# cv2.putText(frame, status, bottomLeftCornerOfText, font, fontScale, fontColor, lineType)
# with open('emotion.txt', 'a') as fp:
# fp.write('{},{},{},{},{},{},{}\n'.format(time.time(), angry, fear, happy, sad, surprise, neutral))
f = frame
else:
f = cv2.flip(frame, 1)
key = cv2.waitKey(20)
if key == 27: # exit on ESC
break
vc.release()
cv2.destroyWindow("exit on ESC")
def facing(self):
detect_face.detect_faces('faces.png')