File size: 1,681 Bytes
c91fae7 c50a96c c91fae7 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 |
import cv2
def discover_square_crop_points(points, image, step=3, maximum_step=30):
x = points["x"]
y = points["y"]
w = points["w"]
h = points["h"]
if step >= maximum_step:
return {"initial_x": x, "initial_y": y, "final_x": x + w, "final_y": y + h}
img_height = image.shape[0]
img_width = image.shape[1]
crop_offset = ((w + h) / 2) / step
initial_x = x - crop_offset
initial_y = y - crop_offset
final_x = x + w + crop_offset
final_y = y + h + crop_offset
if initial_x < 0 or initial_y < 0 or final_x > img_width or final_y > img_height:
return discover_square_crop_points(points, image, step+1, maximum_step)
initial_x = int(initial_x)
initial_y = int(initial_y)
final_x = int(final_x)
final_y = int(final_y)
print(f"step: {step}")
return {"initial_x": initial_x, "initial_y": initial_y, "final_x": final_x, "final_y": final_y}
def crop_faces(cv2_image):
gray = cv2.cvtColor(cv2_image, cv2.COLOR_BGR2GRAY)
face_cascade = cv2.CascadeClassifier(cv2.data.haarcascades + 'haarcascade_frontalface_alt2.xml')
multiScaleValues = [1.01]
for value in multiScaleValues:
faces = face_cascade.detectMultiScale(image=gray, scaleFactor=value, minNeighbors=4, minSize=(512, 512))
if len(faces) > 0:
break
cropped_faces = []
for (x, y, w, h) in faces:
points = {"x": x, "y": y, "w": w, "h": h}
points = discover_square_crop_points(points, cv2_image)
initial_x = points["initial_x"]
initial_y = points["initial_y"]
final_x = points["final_x"]
final_y = points["final_y"]
face = cv2_image[initial_y:final_y, initial_x:final_x]
cropped_faces.append(face)
return cropped_faces |