Question

I am new to image processing and i've been getting my hands dirty with opencv for python for past few weeks. Today I tried to use the default haarcascade XML file for the face detection. Here is my code:

import cv2
import numpy as np

front_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')

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

faces = front_cascade.detectMultiScale(gray, 1.3, 6)
#for (x, y, w, h) in faces:
#    cv2.rectangle(img, (x, y), (x + w, y + h), (0, 255, 155), 3)

print faces

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

Problem here is that it prints out an empty set. And to my surprise when I change the name of the XML file to something like 'aaa.xml', it still produces the same output. Any help would be highly appreciated.

Answer 1

当你说front_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')时，级联文件必须存在于执行程序的同一文件夹中，最佳做法是传递级联文件的完整限定路径，如果你不能找到与opencv捆绑在一起的级联文件，然后您也可以从this link下载.xml个文件。

然后你可以通过传递完整的合格路径加载haarcascde：

front_cascade = cv2.CascadeClassifier('/Users/anmoluppal/Downloads/cascades/haarcascade_frontalface_default.xml')

Answer 2

对于那些试图在Windows conda环境下找到它的人来说，这有助于我找到级联文件。

使用CMD提示符：

dir %CONDA_PREFIX% /s /a /b | findstr /i haarcascade

cannot locate haarcascade default frontal face XML file on windows

2 个答案: