OpenCV的人脸检测功能在一般场合还是不错的。而ubuntu正好提供了python-opencv这个包，用它可以方便地实现人脸检测的代码。
写代码之前应该先安装python-opencv：
具体原理就不多说了，可以参考一下这篇文章。直接上源码。
# face_detect.py
# Face Detection using OpenCV. Based on sample code from:
# http://python.pastebin.com/m76db1d6b
# Usage: python face_detect.py <image_file>
import sys, os
from opencv.cv import *
from opencv.highgui import *
from PIL import Image, ImageDraw
from math import sqrt
def detectObjects(image):
"""Converts an image to grayscale and prints the locations of any faces found"""
grayscale = cvCreateImage(cvSize(image.width, image.height), 8, 1)
cvCvtColor(image, grayscale, CV_BGR2GRAY)
storage = cvCreateMemStorage(0)
cvClearMemStorage(storage)
cvEqualizeHist(grayscale, grayscale)
cascade = cvLoadHaarClassifierCascade(
'/usr/share/opencv/haarcascades/haarcascade_frontalface_default.xml',
cvSize(1,1))
faces = cvHaarDetectObjects(grayscale, cascade, storage, 1.1, 2,
CV_HAAR_DO_CANNY_PRUNING, cvSize(20,20))
result = []
for f in faces:
result.append((f.x, f.y, f.x+f.width, f.y+f.height))
return result
def grayscale(r, g, b):
return int(r * .3 + g * .59 + b * .11)
def process(infile, outfile):
image = cvLoadImage(infile);
if image:
faces = detectObjects(image)
im = Image.open(infile)
if faces:
draw = ImageDraw.Draw(im)
for f in faces:
draw.rectangle(f, outline=(255, 0, 255))
im.save(outfile, "JPEG", quality=100)
else:
print "Error: cannot detect faces on %s" % infile
if __name__ == "__main__":
process('input.jpg', 'output.jpg')
