Python+OpenCV: cv2.imwrite

Question

I'm trying to detect a face and write down area with the face in a separate file. How can I do it? I think that i must use "faces" (you can see this var in code). But how?

from ffnet import mlgraph, ffnet, tmlgraph, imlgraph
import pylab
import sys
import cv,cv2
import numpy
cascade = cv.Load('C:\opencv\data\haarcascades\haarcascade_frontalface_alt.xml')


def detect(image):
 bitmap = cv.fromarray(image)
 faces = cv.HaarDetectObjects(bitmap, cascade, cv.CreateMemStorage(0))
 if faces:
  for (x,y,w,h),n in faces:  
   cv2.rectangle(image,(x,y),(x+w,y+h),(255,255,255),3)
 return image

if __name__ == "__main__":
    cam = cv2.VideoCapture(0)
    while 1:
        _,frame =cam.read()
        frame = numpy.asarray(detect(frame))
        cv2.imshow("features", frame)
        if cv2.waitKey(1) == 0x1b: # ESC
            print 'ESC pressed. Exiting ...'
            break

buran · Accepted Answer · 2019-08-26 20:01:57Z

42

This following code should extract face in images and save faces on disk

def detect(image):
    image_faces = []
    bitmap = cv.fromarray(image)
    faces = cv.HaarDetectObjects(bitmap, cascade, cv.CreateMemStorage(0))
    if faces:
        for (x,y,w,h),n in faces:
            image_faces.append(image[y:(y+h), x:(x+w)])
            #cv2.rectangle(image,(x,y),(x+w,y+h),(255,255,255),3)
    return image_faces

if __name__ == "__main__":
    cam = cv2.VideoCapture(0)
    while 1:
        _,frame =cam.read()
        image_faces = []
        image_faces = detect(frame)
        for i, face in enumerate(image_faces):
            cv2.imwrite("face-" + str(i) + ".jpg", face)

        #cv2.imshow("features", frame)
        if cv2.waitKey(1) == 0x1b: # ESC
            print 'ESC pressed. Exiting ...'
            break

edited Aug 26, 2019 at 20:01

buran

14.4k13 gold badges45 silver badges76 bronze badges

answered Dec 6, 2013 at 14:31

double_g

8367 silver badges9 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

xtluo · Accepted Answer · 2019-08-01 01:02:38Z

6

Alternatively, with MTCNN and OpenCV(other dependencies including TensorFlow also required), you can:

1 Perform face detection(Input an image, output all boxes of detected faces):

from mtcnn.mtcnn import MTCNN
import cv2

face_detector = MTCNN()

img = cv2.imread("Anthony_Hopkins_0001.jpg")
detect_boxes = face_detector.detect_faces(img)
print(detect_boxes)

[{'box': [73, 69, 98, 123], 'confidence': 0.9996458292007446, 'keypoints': {'left_eye': (102, 116), 'right_eye': (150, 114), 'nose': (129, 142), 'mouth_left': (112, 168), 'mouth_right': (146, 167)}}]

2 save all detected faces to separate files:

for i in range(len(detect_boxes)):
    box = detect_boxes[i]["box"]
    face_img = img[box[1]:(box[1] + box[3]), box[0]:(box[0] + box[2])]
    cv2.imwrite("face-{:03d}.jpg".format(i+1), face_img)

3 or Draw rectangles of all detected faces:

for box in detect_boxes:
    box = box["box"]
    pt1 = (box[0], box[1]) # top left
    pt2 = (box[0] + box[2], box[1] + box[3]) # bottom right
    cv2.rectangle(img, pt1, pt2, (0,255,0), 2)
cv2.imwrite("detected-boxes.jpg", img)

edited Aug 1, 2019 at 1:02

answered Dec 13, 2018 at 6:19

xtluo

2,14121 silver badges27 bronze badges

2 Comments

mLstudent33 Over a year ago

for a large number of faces, what I am doing right now is saving all the faces and bounding boxes to lists and then saving these lists of numpy arrays to disk using h5py. Is there a better way?

xtluo Over a year ago

@mLstudent33 I didn't quite get that. You may ask a question on SOF, with more detailed information, e.g. input, output, and what you are doing to accomplish that. Then see anyone could help.

Schmouk · Accepted Answer · 2019-07-03 15:17:38Z

1

wtluo, great ! May I propose a slight modification of your code 2. ? Here it is:

for i, detected_box in enumerate(detect_boxes):
    box = detected_box["box"]
    face_img = img[ box[1]:box[1] + box[3], box[0]:box[0] + box[2] ]
    cv2.imwrite("face-{:03d}.jpg".format(i+1), face_img)

answered Jul 3, 2019 at 15:17

Schmouk

411 silver badge5 bronze badges

3 Comments

Gar Over a year ago

Can you share why your modification is better, or in what cases it would be better?

Schmouk Over a year ago

gar, this modification is a less verbose version of the same code. As such, it might be shorter to read, easier to understand and quicker to process. About the imwrite line (as I can now understand that this was your topic), hasn't it been recommended for a while now to use function format() rather than operator % on strings?

xtluo Over a year ago

@Schmouk thanks for your advice! And my name is xtluo not wtluo. ^_^

Collectives™ on Stack Overflow

Python+OpenCV: cv2.imwrite

3 Answers 3

Comments

2 Comments

3 Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

2 Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related