Search code examples
pythonopencvface-detectionrectangleshaar-classifier

Open CV Python Face Detection


I am looking it to using haar cascades for face detection, so I tried following the tutorial here: http://opencv-python-tutroals.readthedocs.org/en/latest/py_tutorials/py_objdetect/py_face_detection/py_face_detection.html

When I run it though, the image window populates, but the rectangles are not drawn. I checked over the code and everything makes sense to me or if maybe I was using a function that was obsolete, but everything seemed alright.

I am not sure why I cannot get this working. Any suggestions as to what might be wrong? I am using Python 2.7.3.

import numpy as np
import cv2

face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
eye_cascade = cv2.CascadeClassifier('haarcascade_eye.xml')

img = cv2.imread('searchin2.jpg')
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

faces = face_cascade.detectMultiScale(gray, 1.3, 5)

for (x,y,w,h) in faces:
    cv2.Rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
    roi_gray = gray[y:y+h, x:x+w]
    roi_color = img[y:y+h, x:x+w]
    eyes = eye_cascade.detectMultiScale(roi_gray)
    for(ex,ey,ew,eh) in eyes:
        cv2.Rectangle(roi_color,(ex,ey),(ex+ew,ey+eh),(0,255,0),2)

cv2.imshow('img',img)
cv2.waitKey(0)
cv2.destroyAllWindows()

Solution

  • I try to run this script and I got faces: 0 too.

    So I try to find out where is the problem.
    I found out that I have no XML files on my disk.
    There is no info in tutorials that you have to download it seperately and where to get this files.

    But I downloaded OpenCV source code ZIP file, I found haarcascade_frontalface_default.xml in this file and I put it directly in folder with script. Now I have faces: 2 :)

    But now I have also error:

    cv2.Rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
    AttributeError: 'module' object has no attribute 'Rectangle'
    

    I'm still working on it.


    EDIT: It has to be lowercase - rectangle :) Now it works for me :)

    enter image description here

    oryginal image source: wikimedia ?