def predit_face(faces, clf, pca):
"""uses facial recognition model to predict the user's name"
inputs: faces - list of detected faces
clf - facial recognition classifier
pca - Principle component analysis used for classifying faces"""
cropped = followDot.cropImage(frame, faces)
reshaped = manipulateImage(cropped)
X_test_pca = pca.transform([reshaped])
number = clf.predict(X_test_pca)
with open("nameToNumber.txt", "r") as inf:
numberToName = eval(inf.read())
name = numberToName[number[0]]
return name
pitches = []
running = True
while running == True:
# reads camera
ret, frame = camera.read()
# detects faces
faces = face_cascade.detectMultiScale(frame, scaleFactor=1.2, minSize=(20, 20))
# draws rectangle on face
for (x, y, w, h) in faces:
cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 0, 255))
# see if there is a face in the image, enter try if there is
if len(faces) > 0:
cropped = followDot.cropImage(frame, faces)
reshaped = manipulateImage(cropped)
yaw, pitch = ridgeRegression.ridge_predict([reshaped], ridge)
yaws.append(yaw)
pitches.append(pitch)
plt.scatter(yaws, pitches)
previous12values, newYaw, newPitch = updateAverage(previous12values, yaw, pitch)
sendToNeato(pub, twist, newYaw, newPitch)
else: # When no face is detected, stop neato
twist.angular.z = 0
twist.linear.x = 0
pub.publish(twist)
# show the camera with face detection square
cv2.imshow("frame", frame)
