def emotion_detection(self, emotion_model, custom_face):
detected_emotion = dict()
gray_img = functions.detectFace(custom_face, (48, 48), True)
emotion_labels = [
'Kızgın', 'Tiksinti', 'Korku', 'Mutlu', 'Üzgün', 'Şaşırmış',
'Doğal'
]
emotion_predictions = emotion_model.predict(gray_img)[0, :]
sum_of_predictions = emotion_predictions.sum()
mood_items = []
for i in range(0, len(emotion_labels)):
mood_item = []
emotion_label = emotion_labels[i]
emotion_prediction = 100 * emotion_predictions[
i] / sum_of_predictions
mood_item.append(emotion_label)
mood_item.append(emotion_prediction)
mood_items.append(mood_item)
emotion_df = pd.DataFrame(mood_items, columns=["emotion", "score"])
emotion_df = emotion_df.sort_values(
by=["score"], ascending=False).reset_index(drop=True)
for index, instance in emotion_df.iterrows():
emotion_label = "%s " % (instance["emotion"])
emotion_score = instance["score"] / 100
detected_emotion[emotion_label] = emotion_score
return detected_emotion
def age_and_gender_find(self, custom_face, age_model, gender_model):
gender = ""
face_224 = functions.detectFace(custom_face, (224, 224), False)
age_predictions = age_model.predict(face_224)[0, :]
apparent_age = Age.findApparentAge(age_predictions)
gender_predictions = gender_model.predict(face_224)[0, :]
if np.argmax(gender_predictions) == 0:
gender = "Kadın"
elif np.argmax(gender_predictions) == 1:
gender = "Erkek"
analysis_report = str(int(apparent_age)) + " " + gender
self.ui.lbl_yas.setText(str(round(apparent_age)))
self.ui.lbl_cinsiyet.setText(gender)
print("Analysis report: ", analysis_report)
return apparent_age, gender
def loadModelandEmbedding(self, db_path):
global input_shape
model_name = self.model_name
distance_metric = self.distance_metric
employees = []
# liste = list(db.getDersBilgileri(self.dersGenelKod))[0].split(",")
liste = self.liste
print(liste)
print(len(liste))
if len(liste) > 0:
if os.path.isdir(db_path):
for r, d, f in os.walk(
db_path): # r=root, d=directories, f = files
for file in f:
if file.split("_")[0] in liste:
if '.jpg' in file:
exact_path = r + "/" + file
employees.append(exact_path)
if '.png' in file:
exact_path = r + "/" + file
employees.append(exact_path)
print("employees:", employees)
if len(employees) > 0:
if model_name == 'VGG-Face':
print("Using VGG-Face model backend and", distance_metric,
"distance.")
model = VGGFace.loadModel()
input_shape = (224, 224)
elif model_name == 'Facenet':
print("Using Facenet model backend", distance_metric,
"distance.")
model = Facenet.loadModel()
input_shape = (160, 160)
else:
raise ValueError("Invalid model_name passed - ",
model_name)
threshold = functions.findThreshold(model_name,
distance_metric)
tic = time.time()
pbar = tqdm(range(0, len(employees)), desc='Finding embeddings')
embeddings = []
for index in pbar:
employee = employees[index]
pbar.set_description("Finding embedding for %s" %
(employee.split("/")[-1]))
embedding = []
res = functions.detectFace(employee, input_shape)
print(res)
if res is not None:
img = functions.detectFace(employee, input_shape)
img_representation = model.predict(img)[0, :]
embedding.append(employee)
embedding.append(img_representation)
embeddings.append(embedding)
else:
print(f'Resimde Yüz Bulunamadı :{employee}')
continue
df = pd.DataFrame(embeddings, columns=['employee', 'embedding'])
df['distance_metric'] = distance_metric
toc = time.time()
print("Embeddings found for given data set in ", toc - tic,
" seconds")
return df, model, threshold, input_shape
else:
QMessageBox.critical(
self, "Dikkat",
"Sınıf Listesi Yüklenemedi loadModelandEmbedding")
def Camera(self):
global x, y, w, h
ret, img = self.capture.read()
raw_img = img.copy()
if not self.freeze:
faces = self.face_cascade.detectMultiScale(img, 1.3, 5)
if len(faces) == 0:
self.face_included_frames = 0
else:
faces = []
detected_faces = []
face_index = 0
for (x, y, w, h) in faces:
if w > 130:
self.face_detected = True
if face_index == 0:
self.face_included_frames = self.face_included_frames + 1
profile_image_copy = img.copy()
cv2.rectangle(img, (x, y), (x + w, y + h), (67, 67, 67), 1)
cv2.putText(
img, str(self.frame_threshold - self.face_included_frames),
(int(x + w / 4), int(y + h / 1.5)),
cv2.FONT_HERSHEY_SIMPLEX, 4, (255, 255, 255), 2)
self.profile_image = profile_image_copy[int(y):int(y + h),
int(x):int(x + w)]
detected_faces.append((x, y, w, h))
face_index = face_index + 1
# -------------------------------------
if self.face_detected == True and self.face_included_frames == self.frame_threshold and self.freeze == False:
self.freeze = True
self.base_img = raw_img.copy()
self.detected_faces_final = detected_faces.copy()
self.tic = time.time()
self.sayac = 0
if self.freeze:
toc = time.time()
if (toc - self.tic) < self.time_threshold:
if self.freezed_frame == 0:
for detected_face in self.detected_faces_final:
x = detected_face[0]
y = detected_face[1]
w = detected_face[2]
h = detected_face[3]
cv2.rectangle(img, (x, y), (x + w, y + h), self.text_color,
3)
custom_face = self.base_img[y:y + h, x:x + w]
self.age, self.gender = self.age_and_gender_find(
custom_face, self.age_model, self.gender_model)
self.emotion = self.emotion_detection(
self.emotion_model, custom_face)
custom_face = functions.detectFace(custom_face,
self.input_shape)
print("Gerçek Emotion :", list(self.emotion)[0])
self.ui.lbl_duygu.setText(list(self.emotion)[0])
if custom_face.shape[1:3] == self.input_shape:
if self.df.shape[
0] > 0: # Yüz eşleştirilmesi yapılacak yüzlerin olması durumunda işleme
# girmektedir.
img1_representation = self.model.predict(
custom_face)[0, :]
def findDistance(row):
distance_metric = row['distance_metric']
img2_representation = row['embedding']
distance = 1000
if distance_metric == 'cosine':
distance = dst.findCosineDistance(
img1_representation,
img2_representation)
elif distance_metric == 'euclidean':
distance = dst.findEuclideanDistance(
img1_representation,
img2_representation)
elif distance_metric == 'euclidean_l2':
distance = dst.findEuclideanDistance(
dst.l2_normalize(img1_representation),
dst.l2_normalize(img2_representation))
return distance
self.df['distance'] = self.df.apply(findDistance,
axis=1)
df = self.df.sort_values(by=["distance"])
candidate = df.iloc[0]
self.employee_name = candidate['employee']
self.best_distance = candidate['distance']
time_left = int(self.time_threshold - (toc - self.tic))
if self.best_distance <= self.threshold:
self.display_img = cv2.imread(self.employee_name)
if self.sayac < 1:
self.getIformation(self.employee_name)
self.sayac += 1
cv2.rectangle(img, (10, 10), (90, 50), (67, 67, 67), -10)
cv2.putText(img, str(time_left), (40, 40),
cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 255, 255), 1)
self.freezed_frame = self.freezed_frame + 1
self.displayImage(img)
else:
self.face_detected = False
self.face_included_frames = 0
self.freeze = False
self.freezed_frame = 0
self.UI_Ayar()
else:
self.displayImage(img)